Copy and count.
How can heating and cooling a sealed tube find one stretch of DNA among billions, copy it a billion times, and count how much of it the sample held?
The empty wells.
On 4 February 2020 the US Food and Drug Administration authorized, for emergency use, a test for the new coronavirus that the Centers for Disease Control and Prevention had designed in the weeks since the virus's genome sequence was published. Until then, every test in the country had been run in the CDC's own laboratories. The next day the CDC began shipping kits: about 200 for public health laboratories in the United States and a similar number for laboratories abroad, each kit enough for 700 to 800 specimens. Each kit held the primers and probes for four reactions. Two, called N1 and N2, looked for stretches of the new virus's genome. A third, N3, looked for a broader group of related coronaviruses, and the fourth for a human gene, as a check on the specimen and its extraction.
Before a public health laboratory could report a single patient's result, it had to verify the kit on its own instrument, and the CDC expected that to take a few days. Every run included no-template controls: wells holding everything except a template, which should give no signal in any of the 45 cycles. Within days of the first shipments, laboratories reported light in wells that held no template. NPR later reported that the New York City public health laboratory began verifying the kit at once and, within hours, had inconclusive results from two of its reagents. On 12 February the CDC told reporters that some states had found inconclusive results while verifying the kit, and that one reagent was not performing as it should. On 28 February it told reporters that the third reaction, the cause of the inconclusive results, could be left out.
When CDC scientists later studied material from that first lot, false signals appeared in 97 % of the N3 no-template wells made with the kit's reagents, between cycles 34 and 39. The test called a target detected if its light crossed the detection line before cycle 40, so a well with no virus in it read as detected.
The faults that the CDC's scientists identified lay in the design of N3 and in how the reagents were handled in bulk before they were dispensed into kits, not in the instruments. Each instrument had heated and cooled its wells on schedule, read their light and reported a cycle number. Each reaction had done what every PCR does: it copied whatever its primers started, doubling it cycle after cycle until the light crossed a line. In the N3 wells, the reverse primer and the probe had paired with each other, and the reaction copied that pair. The episode shows the whole subject in a few wells: how two short primers find one sequence among billions, how heating and cooling copy it, how the cycle at which a well's light crosses a line counts it, and what evidence a test needs before anyone trusts its count.
Before we start.
A PCR instrument changes the temperature of small sealed tubes on a schedule, and real-time instruments also watch them glow. Everything else happens in the liquid. This guide follows that liquid and the machine around it, and asks one question: how can heating and cooling a sealed tube find one stretch of DNA among billions, copy it a billion times, and count how much of it the sample held?
A few ordinary words have a narrower meaning here. The template is the DNA being copied, and the target is the stretch of it that a test looks for. A primer is a short strand of DNA, made to order, that marks where copying starts; a probe is a similar strand that carries a dye and reports when its target is copied. DNA melts when its two strands separate, not when a solid turns to liquid. A well is one of the small tubes of a laboratory plate, which usually holds 96 of them in a grid. The guide's figures are also called plates, so the text says "Plate 12" for a figure and "a 96-well plate" for the plastic. An assay is the chemistry and procedure of one test: its primers, probes and reaction mix, and the temperatures and times it runs at.
The short answer, which the chapters build, is this. The ingenuity is in the chemistry; the instrument's job is to make the chemistry repeatable. Heat can pull DNA's two strands apart without breaking either, because the bonds between the strands are far weaker than the bonds along each strand, and cooling lets the strands pair again. Two primers pair stably only where their sequence matches, at the temperature the reaction uses, and only the stretch between two primers that face each other is copied over and over: that is how one sequence is found. A heat-stable enzyme builds each new strand, and every copy becomes a template in the next cycle, so the target roughly doubles each time: that is how it is copied a billion times. A dye or a probe gives off light in proportion to the copies, and the cycle at which the light crosses a fixed line shows how much target there was at the start: that is how it is counted. The instrument makes the temperature exact in every well and reads the light at the same moment of every cycle. For a test sold as a kit for patient results, the instrument, the reagents and the software are designed, verified and authorized together, as one system. A laboratory that develops its own test must establish that test's performance itself (Chapters 17 and 25).
Underneath all of this sits one physical limit: in each cycle, the number of target molecules can at most double. Read forward, the limit says that thirty perfect cycles turn one molecule into about a billion. Read backward, it says that the number of cycles a sample needs to reach a fixed amount of product shows how many doublings it was short of that amount, and so how much it held at the start. Copying and counting are one limit seen from its two ends.
How to read this guide
The guide has nine parts. Its chapters are numbered through the whole book, and each begins with a question; within a part, it is usually one that the chapter before it leaves open. The first four parts answer the central question. The next two show how the instrument and the test are engineered and proven, and the last three look at what may replace the cycle, who makes PCR in 2026, and how it is regulated as of October 2026. A reader who buys or evaluates instruments and tests can go from the short answer to Chapters 9, 10, 13, 15, 17 and 21, and one who builds them will want all of Part V and Chapters 22, 25 and 26 as well. Plates have their own numbers, so that any one can be cited by itself. Four kinds of box recur:
Blue edge. Carries the structural point of a section, or the chapter's solved example.
Orange edge. Names a common misreading, a trap, or the limit of a claim.
Green edge. Maps the idea onto the reader's own work: in instrument and assay design, at the bench, in verification, or in a regulatory file.
- Each chapter closes with what it established, in four lines.
The mathematics goes no further than doubling, powers of two and of ten, percentages and ratios. Where a logarithm would normally appear, the guide counts doublings or tenfold steps instead, and the plates draw them. Every chapter has one solved example that works through real numbers step by step and states its assumptions. A plate that is schematic rather than drawn from data says so. The guide uses US spelling.
Inside the plates, color is a legend and never decoration. Template DNA from the sample, the copies made in the tube, primers and probes, RNA, and heat each have one color throughout the guide. The measured readout, the fluorescence and everything computed from it, is always blue, and one orange mark in each plate points at the detail that matters most. A key strip under every plate lists only the colors that plate uses.
This is a reference for understanding, not a laboratory protocol, an engineering specification, a regulatory opinion or clinical advice. Regulatory status, standards editions and product details are stated as of October 2026 and will date. Where a figure is a manufacturer's claim, or comes from a study written by a manufacturer's staff, the text says so. Named instruments and tests are examples of a principle, not recommendations.
+ Part I · The molecule
What PCR copies, and why copying is needed.
The CDC's own scientists showed that their test for SARS-CoV-2 detects about five copies of its target in a reaction, while a band on a stained gel needs roughly two billion copies of a short product or more. Part I sets out that gap in numbers. It then describes the structure of DNA that lets heat pull the two strands apart without breaking either one, and it takes stock of what a cell uses to copy its DNA and how little of that a tube needs. Nothing here assumes more than school biology.
Why copy DNA at all?.
+ The questionA swab already holds the virus's genetic material. Why can no ordinary instrument read it until it has been copied a billion times?
The information is there, in too few molecules
A nasal swab from someone with a respiratory infection carries the answer to a diagnostic question in a very specific form: a sequence of letters in the genome of the virus, written in a molecule a few nanometers wide. The sequence identifies the virus, and it is present in the sample from the moment the swab is taken. The information is not missing, but it is carried by so few molecules that no ordinary instrument can register it.
The scale is easiest to see by weighing a target, the stretch of sequence that a PCR test sets out to copy. One base pair of double-stranded DNA has an average mass of about 650 daltons. A dalton is the unit of mass for atoms and molecules, roughly the mass of one hydrogen atom. A typical target is a stretch of 200 base pairs, so it weighs about 0.2 attograms, or 0.2 billionths of a billionth of a gram. The first complete sequence of a human genome was published in 2022, and it runs to about 3.05 billion base pairs. Even a whole human genome weighs only about 3.3 picograms, which is why a nanogram of human DNA, an amount too small to see, contains only about 300 copies of any one gene.
One base pair of double-stranded DNA has an average mass of about 650 daltons, and one attogram is about 600,000 daltons, so one base pair weighs about 0.00108 attograms. A target of 200 base pairs weighs 200 × 0.00108, or about 0.22 attograms. A haploid human genome, one set of chromosomes, has 3.055 billion base pairs, so it weighs 3.055 billion × 0.00108, or about 3.3 million attograms. A picogram is a million attograms, so the genome weighs about 3.3 picograms. A nanogram is 1,000 picograms, so a nanogram of human DNA holds 1,000 ÷ 3.3, or about 300 haploid genomes. That nanogram therefore carries about 300 copies of a gene that occurs once per genome.
A diagnostic test rarely has even that many copies of its target to work with. The CDC's real-time test for SARS-CoV-2 is the case this guide returns to. In that test, each 20-microliter reaction receives 5 microliters of purified extract, the genetic material recovered from the swab, and the CDC's own scientists showed that it detects about five copies of the target per reaction, measured with RNA transcripts made in the laboratory. Five molecules of a few hundred base pairs weigh about one attogram, a millionth of a millionth of a microgram.
What an instrument needs to see
A detector needs far more. The classic way to see DNA after PCR is to run it through a gel, which separates DNA molecules by size, and to stain it with a fluorescent dye such as ethidium bromide. DNA of one size then shows as a band. A visible band needs roughly half a nanogram to five nanograms of DNA, which for a 200-base-pair product is between about 2 × 10⁹ and 2 × 10¹⁰ copies. In real-time PCR the instrument watches fluorescence build up during the reaction, and its detection line sits at a similar order of magnitude: working backward from published figures places that line at around 10¹⁰ to 10¹¹ copies of product per reaction.
So the gap between what is in the tube and what an instrument can register is about a billionfold, and often more. Five copies are not slightly below the detection limit. They are nine or ten orders of magnitude below it. No refinement of the detector closes a gap of that size at a price a clinical laboratory can pay. Concentrating the sample cannot close it either: putting more of the extract into the reaction gains a few-fold, nowhere near a billionfold, and no concentration step creates molecules that were never collected.
The solution is to make more of the molecule itself. If the target can be copied, and the copies copied, the sequence that was present in five molecules can be made present in ten billion, and an ordinary fluorescence detector or a stained gel can then see it. Making many copies of a sequence in this way is called amplification. Copying has a second advantage that no signal amplifier has: each copy carries the same sequence as the original, so the product can be checked for size, sequenced, or read by a probe, a short labeled piece of DNA that binds only where its sequence matches. The identity of the target is preserved through the whole process.
A positive result means that a particular stretch of genetic sequence was present in the reaction above the test's limit. It does not, on its own, show that the organism was alive, intact or infectious. A fragment of RNA from a cleared infection carries the same sequence as a whole virus. Chapter 13 returns to what the numbers from a PCR test can and cannot say about infectiousness.
Two demands on any copier
Copying sounds simple until two demands are written down. The first is selectivity, copying the target and nothing else. A swab contains human DNA from shed cells, DNA and RNA from the bacteria that live in the nose, and possibly the virus. A human cell alone contains two sets of chromosomes, about six billion base pairs of DNA. The copier must reproduce one stretch of a few hundred letters that belongs to the target and leave everything else alone, or the product will be dominated by whatever was most abundant, which is almost never the target.
The second is speed of multiplication. If a process made one new copy of each original molecule per round, thirty rounds would turn five targets into 155 molecules. That is a linear increase, and it never approaches a billion. To cross nine orders of magnitude in a practical number of rounds, the copies must themselves become templates, strands that are read to make further copies, so that the count multiplies rather than adds. A process in which each round doubles the count reaches a billion from one molecule in thirty rounds, and from five molecules in twenty-eight.
The paradox
The polymerase chain reaction, PCR, meets both demands, and it does so with an instrument that looks almost trivially simple. That instrument is a thermal cycler, often called a PCR machine: a metal block that holds the reaction tubes and changes their temperature on a programmed schedule. A real-time PCR instrument adds a light source and a detector, so that it can watch the reaction as it runs. Nothing in either instrument touches the DNA. No mechanism grips a molecule, no sensor finds a sequence, and no step lets the instrument decide what to copy.
The chemistry in the tube resolves this paradox. The selectivity comes from two primers, short pieces of designed DNA that bind only where their sequence matches (Chapter 5). The multiplication comes from a polymerase, an enzyme that copies DNA, and from a cycle of temperatures that lets every copy serve as a template in the next round (Chapters 4, 6 and 7). The counting comes from watching how many cycles it takes for the copies to become visible (Chapters 11 to 14). The instrument supplies exact temperatures and a precisely timed reading of light, and Chapters 9, 10 and 15 to 19 show that doing those two things well is a serious engineering problem.
Sensitivity is set by the number of target molecules that reach the reaction, not by their concentration in the specimen, the material collected from the patient. A requirement such as "detect 1 copy per microliter of extract" becomes about 5 copies per reaction when 5 microliters of extract go into each well, the small chamber that holds one reaction. That number, a handful of molecules, is what the chemistry, the optics and the statistics of sampling must handle. Chapter 6 shows why no PCR can be reliably positive below about three copies per reaction.
The CDC test appears in almost every chapter. It is a well-documented example of a real-time RT-PCR test, in which the virus's RNA is first copied into DNA (Chapter 20). Its primers and probes were published, its instructions for use and its authorization letters are public, and it ran on a named instrument. Its first weeks, in February 2020, show what goes wrong when design, manufacture and quality control fall short under time pressure. The opening pages told part of that story, and Chapter 22 tells it in full.
- The genetic information in a sample is present in very few molecules: a 200-base-pair target weighs about 0.2 attograms, and a test may need to detect about five copies.
- Gels and fluorescence detectors need roughly 10⁹ to 10¹¹ copies, so the gap is about a billionfold, and only copying the molecule itself closes it at a price a clinical laboratory can pay.
- A useful copier must be selective, copying one stretch among billions of letters, and exponential, with copies becoming templates.
- PCR meets both demands with chemistry, and the instrument only sets temperatures and reads light, but doing those two things exactly is an engineering problem.
What heat can do to DNA.
+ The questionWhat is it about DNA's structure that lets heat pull the two strands apart without breaking either one?
Two kinds of bond
DNA is two long chains wound around each other. Each chain, or strand, is a backbone of sugar and phosphate groups joined end to end by covalent bonds, and from every sugar a base sticks out sideways: adenine, thymine, guanine or cytosine, the A, T, G and C of the genetic code. The sequence of bases along one strand is the information. The two strands run in opposite directions, and they face each other base to base, so that the double helix looks like a ladder whose rungs are pairs of bases. A two-stranded molecule of this kind is called a duplex.
The two kinds of connection in this structure differ enormously in strength, and PCR depends on that difference. Along each strand, the links between one sugar and the next phosphate are covalent bonds, shared electrons that need a chemical reaction to break. Between the strands there are no covalent bonds at all. The strands are held together by many weak interactions, each of which ordinary thermal motion can disrupt.
Two of those weak interactions matter. The first is base pairing: A sits opposite T, held by two hydrogen bonds, and G sits opposite C, held by three. The shapes and hydrogen-bonding patterns of the bases fit only in these combinations, which is why one strand determines the sequence of the other. The second is stacking: the flat bases lie on top of one another along the helix like a pile of coins, and the attraction between neighboring pairs in the stack is what holds the duplex together most firmly.
A common account says that hydrogen bonds hold the two strands together, and that DNA rich in G and C separates into single strands, or melts, at a higher temperature because a G–C pair has three hydrogen bonds instead of two. When the two contributions were measured separately, the result was different. Stacking between neighboring base pairs provides most of the duplex's stability and nearly all of its dependence on temperature and salt. Base pairing itself contributes little, and for an A–T pair it is slightly destabilizing. About half of the extra stability of G–C-rich DNA comes from stacking. The useful division is this: hydrogen bonding decides which base pairs with which, and stacking decides how firmly the paired strands stay together.
Melting
Heat is random molecular motion. As a DNA solution is warmed, the motion grows until it overwhelms the weak forces between the strands. The covalent backbone is far stronger and stays untouched, so the duplex comes apart into two intact single strands. This separation is called melting or denaturation, though nothing turns from solid to liquid. The temperature at which half the DNA is single-stranded, the midpoint of melting, is the melting temperature, written Tm.
Melting is not gradual across all temperatures. Because each base pair's stability depends on its neighbors in the stack, a region that starts to open makes the next region easier to open, and long DNA comes apart over a narrow range of a few degrees. The Tm itself depends on composition and on the solution. A classic study measured long genomic DNA in a fairly salty buffer, 0.2 molar sodium, and found that the Tm rises in a straight line with G–C content: about 69.3 °C plus 0.41 °C for every percentage point of G and C.
For long DNA in 0.2 molar sodium, the melting temperature is about 69.3 °C plus 0.41 °C for every percentage point of G and C. For DNA with 40 % G and C, the increase is 0.41 × 40 = 16.4 °C, so its melting temperature is 69.3 + 16.4, or about 85.7 °C. For DNA with 60 % G and C, the increase is 0.41 × 60 = 24.6 °C, so its melting temperature is 69.3 + 24.6, or about 93.9 °C. In the buffer of a PCR the salt is lower and magnesium is present, so the exact values differ. The lesson holds all the same: almost all natural DNA is fully single-stranded a few degrees above 90 °C, and a very G–C-rich target may need 98 °C.
The solution matters as much as the sequence. The two backbones carry negative charges that repel each other, and positive ions in the solution shield those charges, so more salt means a more stable duplex and a higher Tm. Raising sodium from the tens of millimolar to one molar can shift the Tm of a short DNA strand by as much as 20 °C, and magnesium, which every PCR contains, has an even larger effect per unit of concentration. The concentration of the strands themselves also moves the Tm of short pieces of DNA by several degrees. A melting temperature is therefore a property of a sequence in a particular solution, not of the sequence alone.
These melting temperatures are why PCR denatures at 94 to 98 °C. The temperature is chosen to be safely above the Tm of the target in the reaction buffer, so that every duplex opens in a few seconds. It is also kept low enough for the polymerase in the tube to survive as long as possible, and Chapter 7 quantifies that balance.
Cooling puts the strands back
Melting is reversible. When the solution cools, single strands collide. Two stretches of DNA are complementary when each base in one can pair with the base opposite it in the other, A with T and G with C. When two complementary strands meet, the base pairs form, the stack reassembles and a duplex reappears. This re-pairing is called annealing or renaturation, and its speed depends on how often the right partners meet, which depends on their concentration.
PCR uses annealing deliberately. When the reaction cools after denaturation, two kinds of pairing compete. The long original strands can find each other again, or the primers can find their matching sites on the single strands. The primers are the short designed pieces of DNA of Chapter 1, and they are added in vast excess. Chapter 3 shows that the primers outnumber the template strands by about two hundred million to one, so the primers almost always win.
Temperature also decides how perfect a match must be. A duplex with a mismatched base pair in it is less stable than a perfect one and melts at a lower temperature. At a carefully chosen annealing temperature, a primer that matches its target perfectly stays bound, while one bound to a near-match comes apart from it. The strictness that the annealing temperature imposes on a match is called stringency, and it is the first of PCR's defenses against copying the wrong sequence. Chapter 5 shows how it is used.
What heat costs
The backbone survives 95 °C, but heat still damages DNA chemically, though slowly. The most common change converts cytosine into uracil, which the polymerase then reads as thymine. Two scientists at New England Biolabs, a polymerase supplier, measured this damage in a single-molecule sequencing study and found about 1.4 changes per million bases per cycle. A 100-base target copied over 40 cycles offers 4,000 chances for that damage, one for each base in each cycle, so it carries about 0.006 changes on average, and fewer than one molecule in a hundred carries a heat-induced change. That damage is irrelevant to detecting a virus, but it matters to anyone using PCR to prepare DNA for sequencing.
The larger cost falls on the polymerase. Every second spent at denaturing temperatures shortens its working life, and the loss accelerates sharply above 95 °C. This loss is the reason the top temperature of the cycle is not simply set as high as possible, and it is also why an instrument that overshoots its set point is not harmless.
An assay sets its denaturing temperature as a compromise between opening every target duplex and preserving the enzyme. An instrument that runs a degree cold at the top risks incomplete denaturation and late or absent amplification. One that runs hot shortens the enzyme's working life. Chapter 10 shows that both errors occur in real instruments and how they are measured.
- Each DNA strand has a covalent backbone, and the two strands are held together only by weak interactions, mainly base stacking, while hydrogen bonding decides which bases pair.
- Heat separates the strands without breaking them, the melting temperature depends on G–C content, salt, magnesium and strand concentration, and almost all natural DNA is single-stranded a few degrees above 90 °C.
- Cooling lets complementary strands pair again, and because mismatched pairs are less stable, the annealing temperature sets how exact a match must be.
- Heat has costs: slow chemical damage to DNA and, more important, loss of enzyme activity, which grows quickly above 95 °C.
What a tube can borrow from a cell.
+ The questionA cell uses more than a dozen proteins to copy its DNA. How few can a tube manage with?
How a cell copies its DNA
A dividing cell copies its DNA with a team of proteins, and a PCR tube manages with just one of them. In the cell, the team works at a moving structure called the replication fork, and it copies the entire genome every time the cell divides. A helicase travels along the double helix and unwinds it, separating the two strands. Proteins that bind single strands keep them from pairing again. A primase lays down a short RNA primer on each exposed strand. A DNA polymerase then extends each primer by adding nucleotides, the units of a strand, one by one, and a sliding clamp keeps the polymerase attached. Other enzymes relieve the twisting strain ahead of the fork and replace the RNA primers with DNA, and a ligase seals the gaps.
At the center of this machinery is the polymerase, and PCR has to follow its rules. A DNA polymerase cannot start a new strand from nothing: it can only add to the free end of an existing strand that is paired to a template, the end chemists call the 3′ end. It reads the template one base at a time and adds the nucleotide that pairs with it: A opposite T, G opposite C. And it works in one direction only, extending each new strand from its 5′ end toward its 3′ end. Lengthening a primer along its template in this way is called extension. The polymerase's building blocks are deoxynucleoside triphosphates, or dNTPs, one for each base, and each one added releases a small molecule called pyrophosphate. The reaction also needs magnesium ions to proceed.
- It needs a primer: a short stretch already paired to the template, ending in a free 3′ end.
- It copies the template by pairing: each base added is the complement of the base opposite.
- It extends in one direction only, toward the 3′ end of the new strand.
Everything about how PCR is designed follows from these three rules, from the orientation of the primers to the fact that only the stretch between them is copied.
What the tube keeps and what it replaces
PCR keeps the polymerase and replaces almost everything else with something simpler. Heat replaces the helicase. Chapter 2 showed that raising the temperature above the melting temperature separates the strands without breaking them, so no unwinding enzyme is needed. No proteins are needed to hold the strands apart either, because at 95 °C they stay apart on their own.
Synthetic DNA replaces the primase. Instead of an enzyme that lays down RNA primers wherever replication starts, PCR uses two short pieces of DNA, about 18 to 30 bases long, designed to match the two ends of the stretch to be copied. These primers do two jobs at once. They give the polymerase the starting point it needs, and because they bind only where their sequence matches, they decide which stretch is copied.
The rest of the cell's team is unnecessary. A PCR target is short, typically tens to a few thousand bases, so there are no long strands to untangle and no fragments to join. The polymerase works on its own, and in most diagnostic PCR it also works without the proofreading that cells use to correct errors, a choice that Chapter 7 explains.
What remains is a short list of ingredients, which a manufacturer's protocol gives for a typical 50-microliter reaction with Taq polymerase, a polymerase that survives the heat of the cycle (Chapter 7). The protocol recommends each of the four dNTPs at about 200 micromolar, magnesium chloride at 1.5 to 2 millimolar and each primer at 0.1 to 0.5 micromolar, and it also lists about 1.25 units of polymerase, a buffer to hold the pH and salt, and the template. For the template, the protocol gives anything from a nanogram to a microgram of human DNA, or from 1 picogram to 10 nanograms of a small viral genome or of a plasmid, a small ring of DNA that bacteria carry.
A ready-made mixture of the polymerase, the dNTPs, the magnesium and the buffer is called a master mix, and the primers, any probes and the sample are added to it. Magnesium needs care, because its balance is easy to upset. The polymerase is inactive without enough free magnesium, and too much makes it less accurate and encourages wrong products. The dNTPs bind magnesium, and so do chelators, chemicals that hold metal ions tightly. A sample can carry chelators in, such as EDTA from a blood tube or citrate. What matters is the free magnesium left over, which is one reason a reaction that works with clean DNA can fail with a crude sample.
The primers win by numbers
When a reaction cools after denaturation, each single template strand can either pair with its original partner or bind a primer. The primers win because there are so many of them.
Each primer is at 0.2 µM, which is 0.2 picomoles per microliter, so a 50 µL reaction holds 0.2 × 50 = 10 picomoles of it. A picomole is about 6.0 × 10¹¹ molecules, so the reaction holds about 6 × 10¹² molecules of each primer. The template is 100 ng of human DNA, which at 3.3 picograms per genome is about 30,000 genome copies (Chapter 1), and a target that occurs once per genome has one site in each copy. Each primer therefore outnumbers each target site by about 6 × 10¹² ÷ 30,000, or 2 × 10⁸: two hundred million to one. Each dNTP at 200 µM is 200 picomoles per microliter, or 10,000 picomoles in 50 µL, which is about 6 × 10¹⁵ molecules, enough for far more product than any reaction ever makes (Chapter 8).
At those odds the template strands almost never find each other before a primer finds them. The excess also means that a normal reaction does not use up its primers, a point that matters when Chapter 8 asks why a reaction eventually stops.
The idea that made it a chain reaction
None of the ingredients was new in the early 1980s. Heating DNA to separate its strands had been studied for decades, and chemists had used DNA polymerases to extend short synthetic primers on templates before. Kary Mullis was a chemist making synthetic DNA at the Cetus Corporation in California, and most accounts date his idea to 1983. What he conceived was an arrangement: two primers, one on each strand, pointing toward each other across the target, with the steps of separating, priming and extending repeated over and over. In that arrangement the new strand made from one primer contains the binding site for the other primer, so every product becomes a template in the next round. The process works on its own output, which is why it is called a chain reaction, and why it multiplies rather than adds.
Randall Saiki and colleagues at Cetus, Mullis among them, published the first application in Science in 1985: they amplified a 110-base-pair stretch of the human β-globin gene in order to tell the normal gene from the sickle-cell variant. Their reaction used a polymerase from the bacterium E. coli and ran 20 cycles, with extension at 30 °C, and because the enzyme was destroyed at the denaturing temperature, fresh enzyme had to be added by hand at every cycle. Each cycle copied about 85 % of the target, and twenty of them amplified it about 220,000-fold. Mullis received half of the 1993 Nobel Prize in Chemistry for the invention, and the other half went to Michael Smith for oligonucleotide-based site-directed mutagenesis.
PCR is sometimes credited with a single innovation: putting heat in place of the cell's unwinding enzyme. Heat denaturation was decades old, and the enzyme in the first experiments came from bacteria, not from the cells being tested. The invention was a design that combines two primers facing each other with repeated cycles. That design makes copying exponential and confines it to one chosen stretch.
For an instrument or test developer, the reaction mix is not a commodity poured in at the end. Magnesium, dNTP and primer concentrations, the enzyme and its activation step, and the buffer all set how fast and how specifically the reaction runs, and so they also set the temperatures and hold times that the instrument must deliver. Chapter 17 shows that regulators treat the mix, the instrument and the software as one system for the same reason.
- A cell copies DNA with a team of proteins, and at its center is a polymerase that needs a primer, copies by base pairing and extends in one direction only.
- PCR keeps the polymerase, replaces the unwinding enzyme with heat and the primase with two synthetic DNA primers, and needs only dNTPs, magnesium, a buffer and the template besides.
- Primers outnumber each template site by about two hundred million to one, so they almost always bind before the template strands can pair again.
- Mullis's contribution was the arrangement of two primers facing each other and repeated cycles, which turns copying into a chain reaction.
+ Part II · The cycle
How heat, primers and an enzyme copy one stretch a billion times.
In the first published application of PCR, in 1985, the tube was opened after every heating step and fresh enzyme was added by hand, twenty times in a twenty-cycle run. With a heat-stable enzyme, the tube now stays sealed, and the instrument has one control: temperature. This part shows how a repeating sequence of three temperatures runs three different processes and how two primers select one stretch of DNA. It then follows doubling to a billion, explains why the enzyme had to survive repeated heating to 95 °C before the process could be automated, and shows why every reaction eventually stops.
One knob, three steps.
+ The questionIf nothing can reach into the sealed tube, how can temperature alone run three different steps?
Three processes, three temperature dependences
Once a PCR tube is sealed, nothing more is added to it and nothing is taken out. The instrument cannot stir it, cannot add enzyme at the right moment, cannot remove a product. It can only change the temperature of the metal block the tube sits in. Yet a PCR needs three distinct things to happen, in order, in every cycle: the strands must separate, the primers must bind, and the polymerase must extend them.
What makes this possible is that each of the three processes responds to temperature differently. Strand separation happens only above the melting temperature of the target, so it needs the tube hot, near 95 °C. Primer binding needs the opposite: the short primers form stable duplexes with their targets only below their own melting temperature, typically somewhere around 55 to 65 °C. Extension needs the polymerase to be active, and a polymerase from a heat-loving bacterium works fastest at around 75 to 80 °C, slower at lower temperatures and hardly at all near room temperature. Because different temperature ranges switch the three processes on and off, changing a single variable in the right sequence can run them one after another. Temperature is a switch with three positions.
Denature
The cycle starts hot. At 94 to 98 °C the target duplexes open into single strands (Chapter 2). Once the liquid in the tube is at temperature, a short target separates within seconds, so the denaturing step of each cycle in a modern diagnostic test is brief: the CDC test specifies 3 seconds at 95 °C. The first denaturation of a run is longer, typically 2 minutes, for two reasons. First, long genomic DNA takes longer to open completely the first time, and so does DNA still packed with proteins from the sample. Second, many enzymes are supplied in a blocked form that only an initial period of heat releases, a feature Chapter 5 explains.
Every second spent hot costs enzyme activity, so the denaturing step is kept as short as the target allows. Short steps put a demand on the instrument: it must bring the liquid, not just the block, to temperature quickly, because a step of 3 seconds is meaningless if the sample needs 10 seconds to reach that temperature.
Anneal
Next the instrument cools the tube to the annealing temperature, where the primers bind. The temperature is chosen relative to the primers' own melting temperature. One manufacturer recommends, for its Taq polymerase, an annealing temperature about 5 °C below the lower of the two primers' melting temperatures. Primer binding is fast because the primers are present in enormous excess, so a few seconds are enough (Chapter 3).
The choice of temperature is a trade-off between yield and specificity, which here means how strictly the reaction copies only the target. Too low, and primers stay bound to partial matches elsewhere in the sample, giving products that are not the target. Too high, and even the correct sites are occupied only part of the time, so each cycle copies fewer of the targets. The fraction of targets copied in each cycle is the reaction's efficiency, and an annealing temperature that is too high lowers it. Chapter 5 shows what goes wrong at the low end.
Extend
Then the polymerase extends each bound primer along its template. The rate of extension depends strongly on temperature, and the numbers explain the shape of modern PCR protocols. In 1988 scientists at Cetus, the company that developed Taq polymerase for PCR, published measurements of its speed. They found that it added more than 60 nucleotides per second at 70 °C, about 24 per second at 55 °C, 1.5 per second at 37 °C and a quarter of a nucleotide per second at 22 °C. Scientists at Roche Molecular Systems, which made the enzyme, later characterized it in purified form. They measured about 150 nucleotides per second at 75 °C, close to its optimum.
At 70 °C, Taq adds more than 60 nucleotides per second. A product of 1,000 bases therefore needs about 1,000 ÷ 60, or about 17 seconds, and the common rule of thumb of one minute per thousand bases is deliberately generous. For a target of about 100 bases at 55 °C, where Taq adds about 24 nucleotides per second, the time is about 100 ÷ 24, or about 4 seconds. A short diagnostic target can therefore be extended at the annealing temperature, without a separate hotter step.
For long targets, a separate extension step near 72 °C is still used: a three-step cycle of denature, anneal and extend. For the short targets of diagnostic tests, the anneal and extend steps are usually merged into one. The CDC test specifies 45 cycles of just two steps: 95 °C for 3 seconds, then 55 °C for 30 seconds, during which the primers bind, the polymerase extends them and the instrument reads the fluorescence. The two-step cycle saves a third hold in every cycle, about 45 times per run, while the block travels about the same number of degrees.
Many accounts say 72 °C is the temperature at which Taq works fastest. The purified enzyme is most active at 75 to 80 °C. A temperature around 72 °C is a practical compromise: fast enough for extension, and cool enough that the primers do not fall off before the polymerase has extended them by a few bases, after which the lengthened strand is stable. One enzyme supplier specifies 68 °C in its own protocol. The accuracy of the copy depends mainly on the reaction's chemistry rather than on the extension temperature: on the reaction's pH, its dNTP concentration and the balance of magnesium to dNTPs.
The program is the only instruction
Put together, a PCR run is a temperature program: an initial hold, then a fixed sequence of two or three temperatures repeated 25 to 45 times, sometimes followed by a slow heating to check the product (Chapter 12). Endpoint PCR examines the product only after the last cycle. Its older protocols were written for slower instruments, held each step for 15 to 60 seconds and took one to two hours. A diagnostic real-time test on a fast instrument runs 40 to 45 short cycles in well under an hour, and research systems have run a complete PCR in less than a minute (Chapter 23).
Nothing in the tube tells the instrument how the reaction is going, and in an ordinary thermal cycler nothing in the instrument checks the reaction's progress. The instrument follows its program, and the program assumes that the liquid in every well actually reaches each set point and holds it for the stated time. Whether that assumption holds is the subject of Chapters 9 and 10.
A test's temperature profile consists of its set points, hold times and number of cycles. The profile was fixed when the test was developed, and the test's performance was established with that profile on a particular instrument. Changing the profile changes the test, and so does running the profile on an instrument that delivers it differently. The instructions for the CDC test even named the instrument's run mode: "Standard 7500", not the faster mode the same instrument offered (Chapter 17).
- The instrument controls only temperature, but strand separation, primer binding and extension each respond to temperature differently, so a sequence of temperatures runs them in order.
- Denaturation takes seconds near 95 °C once the liquid is at temperature, and annealing is set a few degrees below the primers' melting temperature, trading yield against specificity.
- Taq's extension rate rises steeply with temperature, so short diagnostic targets are extended during the annealing step, and the CDC test's cycle alternates 95 °C for 3 seconds with 55 °C for 30 seconds.
- The program is the only instruction, and it assumes that the liquid in every well actually reaches each set point.
One stretch out of billions.
+ The questionHow do two short primers pick out one target from billions of DNA letters, and what makes them pick the wrong one?
How rare is a twenty-letter word?
A primer twenty bases long is a word of twenty letters written in an alphabet of four. There are 4²⁰ such words, about 1.1 × 10¹². A human genome has about 3.05 billion base pairs, and read along both strands it offers about 6.1 × 10⁹ places where a twenty-letter word could sit. If genomes were random strings of letters, the expected number of chance matches to any given twenty-letter primer would be about 6.1 × 10⁹ divided by 1.1 × 10¹², or about 0.006. On this arithmetic, a primer designed for a viral sequence would be very unlikely to find a perfect match anywhere in human DNA.
That arithmetic explains why primers work at all, and it also overstates how selective a primer is, for three reasons. First, real genomes are not random. They contain repeated sequences, and some, such as the Alu elements, are present in about a million copies, so a primer that happens to resemble one of them has many places to bind. Second, at a low enough temperature a primer does not need a perfect match to bind: a near-match with one or two mismatches can hold, especially if the mismatches are far from the end the polymerase extends. Third, the sample contains more than human DNA. So the specificity of PCR does not rest on any single primer being unique. It rests on the primers working as a pair, on the annealing temperature, and on the way the polymerase treats the end of a primer.
The pair is what selects
The polymerase extends only from a primer, and only in one direction (Chapter 3). PCR uses two primers designed to bind to opposite strands, at the two ends of the target, each pointing toward the other. When each primer is extended, it produces a new strand that runs across the target and through the binding site of the other primer.
The geometry of the pair is the real filter. A primer that binds by chance at an unrelated site produces a single extended strand. That strand is copied exponentially only if the second primer also binds to it, on the opposite strand and facing back, within the distance the polymerase can cover in one extension step. It is far less likely that both primers find mismatched sites arranged like that, close enough together, than that either finds one alone. A stray binding event usually produces a few strands that grow by one per cycle and never become visible. The true target lies between both primers, so it grows exponentially and soon far outnumbers those strands.
Where the defined product first appears
The geometry also explains a detail that surprises most people when they first follow the cycles one by one. In the first cycle, each primer is extended along a long original strand, and the polymerase simply keeps going past the far end of the target until the step ends. The new strand has one defined end, where the primer started, and one ragged end. Strands of this kind are called long products. In the second cycle, the other primer binds to one of those long products and is extended toward its defined end. Because that template strand ends exactly where the first primer began, the copy stops there, and the result is the first strand of exactly the target's length. In the third cycle, two such exact strands pair into the first double-stranded product with both ends defined by the primers.
From then on, the exact-length product doubles every cycle, while the long products increase by only two per cycle from each original duplex, because only the original strands make them.
The reaction starts from one double-stranded target, at 100 % efficiency. After n cycles it holds 2ⁿ duplexes in all, and 2n of them still contain an original strand or a long product, so 2ⁿ − 2n are of exact length. For cycles 1 to 7 that gives 0, 0, 2, 8, 22, 52 and 114. After 30 cycles it gives 2³⁰ − 60, or about 1.07 × 10⁹ exact-length duplexes, beside just 60 long products. By the time anything is detectable, essentially all of the product is the stretch between the two primers.
Mispriming, primer-dimers and the hot start
Two kinds of wrong product are common, and both start when the temperature is too low for the primers to bind selectively.
The first is mispriming: a primer binds to a site that only partly matches, and the polymerase extends it. Whether that matters depends on which end of the primer the mismatch is near. The polymerase extends only from a 3′ end that is paired to the template, so a mismatch at or very near the 3′ end greatly reduces extension, while a mismatch near the other end of the primer is often tolerated. Primer designers use this deliberately, placing the most discriminating bases near the 3′ end.
The second is the primer-dimer, which forms between the two primers. They are present at enormous concentration, and they bind briefly to each other through a few complementary bases at their 3′ ends. The polymerase then extends both across each other. The result is a short double-stranded product that contains both primer sequences, so it amplifies as efficiently as a real target. It competes with the target for the reaction's resources, and with some detection chemistries it produces signal of its own.
Both errors are most likely before cycling starts. When a reaction is assembled at room temperature, the primers bind loosely to anything partly complementary, and a polymerase that keeps a little activity at room temperature extends those mistakes. In a study published in 1992, scientists at Cetus, the company where PCR was invented, prevented any activity before the first heating step, using a layer of wax that kept the reagents apart until it melted. They found large improvements for targets below about a thousand copies, enough to amplify a single target molecule of HIV in a background of human DNA.
Any method that prevents polymerase activity until the first heating step is called a hot start, and hot-start methods now do the same job in other ways. In one, an antibody binds and blocks the polymerase until heat destroys the antibody, in another a chemical modification of the enzyme is reversed by a long initial heat step, and in a third a short DNA molecule binds the enzyme and lets go when heated. With TaqPath, one of the commercial master mixes named in the CDC test's instructions, the two-minute hold at 95 °C at the start of the test's cycling serves this purpose.
The arithmetic of twenty-letter words shows that a perfect chance match in a genome is unlikely. It does not show that a primer will bind only where intended. Repeated sequences, near-matches at low temperature, primer–primer pairing and activity during set-up all produce wrong products. Specificity is designed through the pair geometry, the annealing temperature, hot start and checks of the sequences, and it is then demonstrated by testing.
Designing for the right target
The design rules that follow from all this are conventions rather than laws, and suppliers state them a little differently. Primers are usually 18 to 30 bases long, with 40 to 60 % G and C and a G or C at the 3′ end to anchor it. The two primers' melting temperatures are kept within about 5 °C of each other, so that one annealing temperature suits both. Runs of four or more identical bases are avoided, and so are sequences that can fold on themselves or pair with the other primer.
For diagnostic tests the target itself is chosen with care. A good target region is conserved across all known variants of the organism, so that no strain escapes detection, and it is absent from the organism's relatives and from human DNA, so that nothing else is detected. Most of these checks are done by computer before any reagent is made. The designer aligns the primers against every available sequence of the target organism to check that all are matched, and against the sequences of related organisms and human DNA to check that none is. During the COVID-19 pandemic, the FDA asked developers to show by such alignments that their primers matched all published SARS-CoV-2 sequences, and to treat any similarity above 80 % between a primer and another organism as a possible cross-reaction to be tested in the laboratory. The agency made both requests in its template for developers seeking emergency authorization.
A test with a probe has at least three oligonucleotides, short strands of DNA made to order, in the same tube. A multiplex test detects several targets in one reaction, so it has many more. Each pair must be checked for complementarity, especially at the 3′ ends, because any two that can pair and be extended can make a product in any reaction, including those with no target. The oligonucleotides for the CDC's original third target, N3, had exactly this flaw: the 3′ end of the probe could pair with the 3′ end of the reverse primer, one of the two primers of the N3 reaction. A no-template control is a reaction with every reagent but no sample, so it should show no amplification. With the reagents of the first authorized kits, false signal appeared in about 97 % of N3 no-template controls, against 0.5 to 2 % for the same sequences from other sources, and it grew more frequent as the reagents aged (Chapter 22).
- A twenty-base primer is unlikely to match a random genome by chance, but repeats, near-matches and primer–primer pairing make chance alone an inadequate defense.
- Two primers on opposite strands, facing each other, are the real filter: only the stretch between them is copied exponentially.
- Exact-length duplexes first appear in cycle 3 and then double each cycle, and after 30 cycles they outnumber the long products by tens of millions to one.
- Mispriming and primer-dimers arise mostly at low temperature and before cycling, and hot start, 3′-end design and computer checks of every oligonucleotide control them.
How one becomes a billion.
+ The questionDoubling thirty times makes a billion. Why do real runs need closer to forty cycles?
Products become templates
In each cycle, every target strand in the tube can bind a primer and be copied, and every new strand contains the binding site for the other primer, so in the next cycle it is a template too. If every target is copied in every cycle, the count doubles: 1, 2, 4, 8, 16. Ten cycles give 2¹⁰ = 1,024, twenty give about a million, thirty give 2³⁰, about 1.07 × 10⁹, and forty give about 1.1 × 10¹². This arithmetic is behind the claim that PCR makes a billion copies in thirty cycles, and it is the forward reading of the fact stated at the start of this guide: a cycle can at most double the number of targets.
Doubling is the ceiling, and real reactions sit below it. The fraction of targets actually copied in a cycle is the efficiency (Chapter 4). At 100 % efficiency every target is copied and the count doubles. At 90 % efficiency, nine targets in ten are copied and the count is multiplied by 1.9. Each cycle multiplies the count by one plus the efficiency, so after a run of cycles the reaction holds the starting number multiplied by that factor once for every cycle.
Efficiency compounds
A shortfall of a few percent per cycle sounds harmless. Compounded over thirty cycles, it is not.
At 100 % efficiency each cycle multiplies the target by 2, so 30 cycles multiply it by 2³⁰, about 1.07 × 10⁹. At 90 % efficiency each cycle multiplies it by 1.9, so 30 cycles multiply it by 1.9³⁰, about 2.3 × 10⁸. Dividing the first figure by the second gives about 4.7, so the reaction at 90 % ends with about 4.7 times fewer copies. In 1985 the first published PCR, by scientists at Cetus (Chapter 3), copied about 85 % of the target per cycle. Its 20 cycles multiplied the target by 1.85²⁰, about 220,000, where perfect doubling would have given 2²⁰, about a million.
Doubling has a shape that is worth learning now, because Part IV depends on it. Plotted on an ordinary scale, an exponential curve hugs zero for most of its length and then shoots up, so the early cycles look like nothing is happening. Plotted on a scale where each step up the axis is a factor of ten, the same curve is a straight line. Perfect doubling climbs one tenfold step every 3.32 cycles, because 3.32 doublings make a factor of ten. At 90 % efficiency the line climbs one step every 3.59 cycles. The steepness of the line is set by the efficiency, and a sample that starts with ten times more target runs along a parallel line, 3.32 cycles ahead at perfect doubling and 3.59 cycles ahead at 90 %.
Efficiency falls short of 100 % for several reasons that recur through this guide. Some strands are not fully separated in a short denaturing step, some primer sites are not occupied at the annealing temperature, some extensions do not finish, and substances carried in from the sample slow the enzyme (Chapter 20). Because every cycle applies the per-cycle factor again, the effect of any difference in efficiency between two reactions grows with every cycle. That matters for counting, because two samples with the same starting amount but different efficiencies reach the detection line at different cycles (Chapter 13).
Where the forty comes from
The number of cycles a test needs is set by three things: how few molecules it must detect, how many molecules the detector needs to see, and the efficiency in between.
A real-time instrument's detection line sits at around 10¹⁰ to 10¹¹ copies of product per reaction (Chapter 1). From a single copy, perfect doubling would reach 10¹⁰ in about 33 cycles. Real reactions do not double perfectly, and they rarely start from exactly one copy. The field's reference for reporting real-time PCR is a set of guidelines called MIQE, short for Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Their 2025 revision gives worked figures: at 90 % efficiency, 10 starting copies cross a typical detection line at about cycle 35 and 3 copies at about cycle 37, while at 70 % efficiency the same inputs cross at about cycles 42 and 45.
A run of 40 to 45 cycles therefore covers the handful of molecules a sensitive test must detect, even with an efficiency below perfect, and little is gained by going further. The CDC test uses 45 cycles and counts a target as detected only if it crosses the line before cycle 40.
Adding cycles beyond about forty does not lower the floor that random sampling sets on detection, and it gives rare side reactions, such as primer-dimers and traces of contamination, more chances to produce a signal. That is one reason tests set a cut-off below the last cycle, as the CDC test does at cycle 40 of 45, and why a late signal in a no-template control invalidates a run.
The floor no instrument can lower
There is a limit below which no number of cycles helps, and it is set not by chemistry but by sampling. When a few microliters of extract are pipetted into a reaction, the number of target molecules that land in the tube is a matter of chance. If the extract contains an average of one target molecule per 5 microliters, some reactions receive none, some one, and some two or three.
The chance that a reaction receives no molecules follows a simple rule, the Poisson distribution, which describes items scattered at random. With an average of one molecule per reaction, 37 % of reactions get none and must be negative however good the chemistry. Each additional molecule in the average multiplies that chance by 0.37 again: about 14 % for an average of two, about 5 % for three. A test's limit of detection is usually defined as the lowest amount of target that it detects in at least 95 % of repeated reactions.
For an average of 1 copy per reaction, the chance that a reaction receives no target is 37 %. For 2 copies it is 0.37 × 0.37, about 14 %. For 3 copies it is 0.37³, about 5 %, and for 5 copies it is 0.37⁵, about 0.7 %. The exact factor is 0.368, which is one divided by the number e, about 2.718. A test that must be positive at least 95 % of the time at its limit of detection therefore cannot have a limit below about three copies per reaction, and the MIQE guidelines state this floor explicitly. The CDC's paper on its test reported detecting about five copies per reaction (Chapter 21).
The same chance governs reactions that do receive molecules. A reaction that receives two molecules instead of four starts one doubling behind, so it crosses the detection line about one cycle later. At very low concentrations, replicate reactions therefore scatter by a cycle or more, for reasons that have nothing to do with the instrument or the chemistry. For this reason, the MIQE guidelines put the lower limit of reliable quantification of a good assay at about ten copies, above the limit of detection.
So a test that is "as sensitive as possible" is one that reliably detects a few copies per reaction, and no extra cycles, brighter dyes or better optics can change that limit. The only way to detect a rarer target is to put more of the sample into the reaction, by extracting from a larger volume or concentrating the extract.
For the instrument, the number of cycles multiplies every second of the cycle. A test of 45 cycles spends 45 times the duration of one cycle, so a saving of two seconds per cycle saves a minute and a half per run. Time to result is one of the requirements that most shape an instrument (Chapter 16). It is set mainly by how quickly the liquid in every well can be moved between two temperatures, cycle after cycle.
- Every copy becomes a template, so at 100 % efficiency the count doubles each cycle, and 30 cycles give 2³⁰, about 1.07 × 10⁹.
- Efficiency compounds: at 90 % per cycle, 30 cycles give about 4.7 times fewer copies than perfect doubling.
- Reaching the detection line from a few copies at realistic efficiency takes about 35 to 45 cycles, which is why tests run 40 to 45.
- Sampling sets a floor: with an average of three copies per reaction, 5 % of reactions receive none, so no PCR can reliably detect fewer.
An enzyme that survives the heat.
+ The questionWhy couldn't PCR be automated until it used an enzyme from a hot-spring bacterium?
The Klenow years
The first PCR experiments used a fragment of DNA polymerase from the bacterium E. coli, called the Klenow fragment. It copied DNA well at moderate temperatures, but it was destroyed by the first heating to separate the strands. So in the first published application of PCR, in 1985, the tube was opened after every denaturation and fresh enzyme was added by hand, twenty times in a twenty-cycle run (Chapter 3). Extension ran at 30 °C, the temperature the enzyme tolerated, and annealing at the same low temperature.
Both features limited the method. Adding enzyme every cycle was laborious and opened the tube twenty times to contamination. And annealing at 30 °C let primers bind to partial matches all over the sample, so the reaction made many products besides the target. The first experiments compensated by detecting the product with a labeled probe, which picked the right product out of the mixture. Cetus engineers built automated prototypes to do the pipetting, and one of them was nicknamed "Mr. Cycle", after a coffee machine. But the real solution was an enzyme that did not need replacing.
A bacterium from Yellowstone
That enzyme came from a microbe isolated twenty years earlier for an unrelated reason. In 1966 the microbiologist Thomas Brock and his undergraduate student Hudson Freeze took samples from microbial mats in Mushroom Spring, in the Lower Geyser Basin of Yellowstone National Park, and the water where they sampled was about 71.5 °C. From one sample they grew a bacterium that thrived in heat, and in 1969 they described it as a new species, Thermus aquaticus. It grows between about 40 and 79 °C, best at about 70 °C.
An organism that lives at 70 °C needs proteins that work at 70 °C. In 1976 the DNA polymerase of T. aquaticus was purified and found to be most active near 80 °C. In 1988 scientists at Cetus, the company where PCR was invented, published PCR with this enzyme, now called Taq polymerase, in place of the Klenow fragment.
Because Taq survived the denaturing step, it could be added once at the start, the tube could stay sealed, and the entire run could be automated with a programmable heater. Because it worked at high temperature, annealing and extension could be done hot, where primers bind only to good matches. The group reported higher specificity and yield, products up to about 2,000 bases long, and amplification of single-copy human genes more than ten-million-fold. A commercial thermal cycler from Perkin-Elmer Cetus had reached the market in 1987, and PCR spread through laboratories within a few years.
Taq is often said to come from microbes living in near-boiling water and to work best at 72 °C. T. aquaticus was isolated from water at about 71.5 °C and grows up to about 79 °C. Its polymerase is most active at 75 to 80 °C (Chapter 4). The enzyme's tolerance of 95 °C is enough to survive brief denaturing steps, not indefinite boiling.
Heat stability has a half-life
Taq is thermostable, not indestructible. In 1993 scientists at Roche Molecular Systems, which made the enzyme, published a characterization of the purified enzyme. They measured how long the enzyme kept half its activity at denaturing temperatures: about 45 to 50 minutes at 95 °C, but only 9 minutes at 97.5 °C. Two and a half degrees cut its half-life by a factor of five.
With a half-life of 45 minutes at 95 °C, Taq's activity halves for every 45 minutes spent there. A classic run of 30 cycles holds 95 °C for 30 seconds in each cycle, which makes 15 minutes hot, or a third of a half-life. A third of a halving leaves 0.5 raised to the power one third, or about 79 % of the activity. At 97.5 °C the half-life is 9 minutes, so the same 15 minutes are 1.7 half-lives and leave about 31 %. The CDC test's program holds 95 °C for 2 minutes at the start and for 3 seconds in each of 45 cycles, about 4.3 minutes in all, and about 94 % of the activity remains. Short denaturing steps spare the enzyme, and they need an instrument that brings the liquid to temperature quickly.
The arithmetic makes concrete a point from Chapter 2: the denaturing temperature is a compromise between opening every target duplex and preserving the enzyme. If an instrument lets the liquid in its wells overshoot 95 °C by a couple of degrees, for a few seconds every cycle, it consumes enzyme life that the assay was not designed to spend, while one whose wells fall short risks leaving strands unseparated. The block itself is often driven past the set point on purpose, but what matters is the temperature of the liquid (Chapter 10). Both errors are invisible unless someone measures them.
No proofreading, and a useful side activity
Cells copy DNA with proofreading polymerases that check each base and remove mistakes. Taq has no such proofreading activity, so it makes more errors. How many depends on how errors are measured, and different methods give results more than tenfold apart for the same enzyme, so comparisons are only meaningful within one study. In 2017 scientists at a company that sells high-fidelity enzymes counted errors by sequencing single molecules, and for Taq they measured about 150 errors per million bases per doubling, against about 5 for the proofreading enzyme Pfu and 0.53 for the company's own engineered enzyme, Q5.
For detection, Taq's error rate hardly matters. For a 100-base target copied through 30 doublings, the average number of errors per final molecule is about 100 × 30 × 150 per million, or about 0.45. But the errors fall at random positions in different molecules, so the population as a whole still carries the correct sequence, and a probe or a primer binds regardless of an occasional changed base elsewhere. When one molecule is picked out of the mixture, to be cloned or to define a sequence, the errors matter, and a proofreading enzyme is used.
Taq has one more activity that turned out to be valuable. As it extends a strand, it degrades any short piece of DNA bound to the template in its path, an activity called 5′ to 3′ exonuclease. In 1991 Taq's developers at Cetus showed that this activity could be used to cut a labeled probe sitting inside the target, generating a signal as the target is copied. Chapter 12 shows how probe cleavage became the most common detection chemistry in diagnostic PCR.
Choosing a polymerase sets several parameters of the whole test at once, starting with the length of the initial activation step: one supplier states that a chemically blocked enzyme may need up to about 10 minutes at 95 °C, and an antibody-blocked one about a minute. The choice also sets the extension rate and therefore the shortest workable cycle, the fidelity of the copy, and whether probe cleavage is available as a detection method. Each choice reaches the instrument as a temperature profile it must deliver.
- Early PCR used an enzyme destroyed at 95 °C, so fresh enzyme was added by hand every cycle and annealing ran at a non-selective 30 °C.
- Taq polymerase comes from a bacterium isolated in a Yellowstone spring at about 71.5 °C, and because the enzyme survives denaturation it could be added once, which made sealed, automated, high-temperature PCR possible.
- Taq's half-life is about 45 to 50 minutes at 95 °C and 9 minutes at 97.5 °C, so short denaturing steps and accurate temperatures preserve it.
- Taq lacks proofreading, which matters for cloning but not for detection, and its 5′ to 3′ exonuclease activity became the basis of probe detection.
Why it doesn't copy forever.
+ The questionIf every cycle doubles the product, why does every run flatten out?
The shape of a real reaction
If doubling continued, forty cycles would turn a single molecule into about a trillion, and fifty into a quadrillion. No reaction does this. Plotted on an ordinary scale, the amount of product in a PCR stays invisibly small for many cycles, then rises steeply, then bends over and levels off. The level at which it stops is called the plateau, and for a standard reaction it corresponds to roughly 0.3 to 1 picomole of product, or about 2 × 10¹¹ to 6 × 10¹¹ molecules.
The plateau does not arrive suddenly. Efficiency declines gradually as product accumulates, so the curve bends rather than breaks: a phase of near-constant doubling, then a phase in which each cycle adds less than the last, and finally a phase in which cycles add almost nothing. That gradual bend is what gives an amplification curve its familiar S shape.
The usual explanation, and what experiments found
The usual explanation is that the reaction runs out of something, primers or building blocks. The numbers from Chapter 3 make that explanation hard to sustain for a normal reaction.
In a 50 µL reaction, each primer at 0.2 µM comes to 10 picomoles, or about 6 × 10¹² molecules. The plateau holds about 2 to 6 × 10¹¹ product molecules, roughly 0.3 to 1 picomole, and each product molecule used one molecule of each primer. At most about a tenth of each primer has therefore been used, so at the plateau nine in ten primer molecules or more are still unused. Each dNTP at 200 µM comes to about 6 × 10¹⁵ molecules. A 100-base-pair product has 200 nucleotides in its two strands, about 50 of each of the four kinds, so the supply could make about 6 × 10¹⁵ ÷ 50, or about 10¹⁴ products, several hundred times more than the plateau. Primer and dNTP depletion can stop a reaction designed with very low concentrations, but they do not explain the plateau of a standard one.
Experiments that looked directly for the cause point elsewhere. The authors of a 1994 study ruled out several causes, each as a sufficient explanation: exhaustion of the enzyme, the dNTPs or the primers, the accumulation of pyrophosphate, and enzyme degradation of the product. Pyrophosphate is the small molecule released with each nucleotide the polymerase adds (Chapter 3). The authors proposed instead that the product strands themselves were responsible. Once product is abundant, the two strands of each product find each other quickly when the reaction cools, and re-pairing with a long, perfectly matched partner competes with binding a short primer. Every strand that re-pairs before a primer reaches it is not copied in that cycle.
The competition shifts as the reaction proceeds, because of how often the molecules meet. The rate at which product strands re-pair depends on how often a product strand meets its complement, so it grows with the square of the product concentration: twice the product, four times the re-pairing. The rate at which primers bind depends on how often a product strand meets a primer, and the primer concentration hardly changes during the reaction. So the fraction of product strands lost to re-pairing in each cycle grows in step with the product itself. While product is scarce, in the early cycles, almost every strand is primed and copied. Once it is abundant, a growing share re-pairs first, and the efficiency falls cycle by cycle.
A study published in 2000 found a second effect of abundant product. The polymerase binds double-stranded DNA in general, not only at primer ends, and as product accumulates the enzyme is increasingly held on product duplexes where it does nothing useful. In that study, short double-stranded DNA fragments inhibited the polymerase completely when they outnumbered its molecules about thirty to one, while the depletion of dNTPs and the loss of enzyme to heat were not the main limits.
Other factors contribute in particular reactions: nonspecific products and primer-dimers competing for enzyme and primers, the gradual loss of enzyme activity to heat, and incomplete separation of strands when the product is very concentrated. The common thread is that the reaction is limited by its own success. The more product there is, the more the product interferes with making more.
The most repeated explanation of the plateau is that the primers and nucleotides run out, and for a standard reaction it is the least supported. Experiments point first to product strands that re-pair instead of binding primers, and to the polymerase being taken up by the accumulated product.
The plateau forgets where the reaction started
The most important consequence of the plateau concerns measurement. The authors of the 1994 study also observed that reactions of the same target started from different amounts reached the same maximum. A sample with a thousand target molecules and a sample with a million both end in roughly the same amount of product, because both are stopped by the same self-limiting effects. The plateau carries almost no information about how much target there was at the start.
Those self-limiting effects depend on the amount of product in the tube, not on how many cycles it took to get there, so a reaction that starts with more target simply reaches the limiting amount sooner. A reaction with a million starting copies is about ten cycles ahead of one with a thousand all the way up the exponential phase, and then both stall at roughly the same height.
The first generation of PCR measured the product at the end of a run, by running it on a gel. That measurement can say whether the target was present and whether the product is the right size, but because of the plateau it cannot reliably say how much target was present. The information about the starting amount is found earlier, in the cycles where the reaction is still doubling. In those cycles, two samples that differ tenfold in their starting amount are separated by a steady gap of a little over three cycles: the sample with more target reaches any given amount of product sooner. Part IV builds on exactly that gap.
For a yes-or-no answer with a size check, an endpoint reaction of 25 to 35 cycles followed by a gel is cheap and adequate. It needs a positive control, a reaction that contains the target at a known level, to show that the run worked. A failure in one sample's own reaction, from a substance in the specimen that slows the enzyme, for example, shows only in a control amplified from that same sample (Chapter 21). For any question of quantity, the reaction must be observed during the exponential phase, which means reading fluorescence in every cycle. The other route is to divide the sample into many small separate reactions, called partitions, so that each partition gives a yes-or-no answer (Chapter 14).
- Every PCR levels off at a plateau, roughly 2 to 6 × 10¹¹ product molecules in a standard reaction, after a gradual decline in efficiency that gives the curve its S shape.
- In a standard reaction the primers and dNTPs are far from exhausted at the plateau, and experiments point to product strands re-pairing and to the polymerase binding accumulated product.
- Reactions started from very different amounts reach the same plateau, so the endpoint says little about the starting amount.
- Quantity must therefore be read in the exponential phase, where a tenfold difference in starting amount appears as a steady gap of a little over three cycles, or counted in many partitions that each give a yes-or-no answer.
+ Part III · The thermal cycler
The box that changes temperature.
In a study published in 2005, researchers tested six new cyclers and found two in which the liquid was still 13 to 20 °C short of the set point when the hold timer of the denaturing step started. On those two instruments a diagnostic test gave no positive results until the step was made hotter and longer. The chemistry assumes that every well reaches each temperature, holds it for the stated time and leaves it on schedule. This part opens the instrument that is supposed to make that true: the path heat takes from the liquid to the room and back, the thermoelectric modules that pump it, and the engineering of speed, evenness and accuracy, including how each is specified, measured and sometimes missed.
Inside the thermal cycler.
+ The questionWhat has to be inside a box whose job is to change temperature on a schedule?
The path heat takes
A PCR reaction is a drop of liquid, typically 10 to 50 microliters, in a thin-walled plastic tube. A thermal cycler moves heat into that liquid and then out again, about forty times per run, as quickly and evenly as possible. Everything inside the box serves that path, and the path is easiest to follow from the sample outward.
Instead of a loose tube, the reaction may sit in a plastic plate, a single tray of small tubes, each called a well. The tube, or the plate's well, sits in a shaped hole in a metal block, which holds 96 or 384 of them. Beneath the block are thermoelectric modules, usually called Peltier modules, which pump heat into the block or out of it depending on the direction of the current through them. Beneath the modules is a heat sink, a large finned piece of metal, and a fan that blows room air across the fins. Above the tubes is a heated lid, a heated cover that presses down on the tube caps. Somewhere in the block is a temperature sensor, and a controller reads it and sets the current in the modules.
Each layer resists the flow of heat and stores some of it, and together those resistances and stores of heat set how the instrument behaves. The plastic wall is a poor conductor, so the liquid always lags behind the block. The block must be massive enough to spread heat evenly across all its wells and light enough to change temperature quickly. The modules must pump enough heat to move the block several degrees per second. The heat sink must carry that heat away to the room. A design that improves one of these usually worsens another, and Chapter 16 shows how such trade-offs are budgeted.
The Peltier module
A Peltier module is a flat array of small blocks of two kinds of semiconductor, connected electrically in series and sandwiched between two thin ceramic plates. When a current flows, it carries heat from one plate to the other. When the current is reversed, the heat flows the other way. The same component therefore heats the block when the current runs one way and cools it when the current runs the other, and because it needs no compressor, no liquid and no moving parts, it suits a compact instrument that must switch direction every few seconds.
The physics comes with losses. The current that pumps heat also generates resistive heat in the semiconductor, and some heat leaks through the module from its hotter plate to its cooler one. A module maker's technical reference states the balance at the face being cooled: the heat removed from it is the heat the current carries, minus half the resistive heat, minus the heat that leaks back into it. The resistive heat works against the module when it cools the block and adds to the heat delivered when it warms it. A PCR block sits above the temperature of the heat sink, so while the block is cooled the leakage runs from the block toward the sink and slightly aids cooling, but the resistive loss dominates.
Cooling is therefore slower than heating in many instruments: one manufacturer specifies its 96-well block at 4.4 °C per second for heating and 2.2 °C per second for cooling. The heat sink must reject both the heat pumped out of the block and all the electrical power put in, so the fan and fins are sized for more heat than the block ever holds.
A Peltier module has no moving parts, but it does wear out. Each swing in temperature expands and contracts its layers by different amounts, because the semiconductor, the solder and the ceramic expand differently with heat. The stress, repeated in every cycle, cracks the joints. A module maker states in its reliability guide that thermal cycling is the operating mode with the highest failure rates, and that failures are often abrupt. The same maker tested cycling-rated modules swung between 30 and 100 °C: every module survived at least 25,000 cycles, and the group's mean time between failures was calculated at 68,000 cycles. In a university study, researchers cycled a commercial module to destruction: its performance had fallen by a fifth after 40,000 cycles, and it failed at 45,000.
An instrument chosen for illustration runs three 45-cycle tests a day. That is 3 × 45 = 135 thermal cycles a day, and 135 × 365 = 49,275, or about 49,000 cycles a year. In the tests above, the maker's modules had a calculated mean time between failures of 68,000 cycles, and the university's module failed at 45,000. One year of this use is therefore of the same order as the cycles to failure in those tests. Both tests swung their modules far wider than a PCR cycle's 40 degrees: 70 degrees in the module maker's test, and more in the university's. A narrower swing stresses the joints less, but the life is still finite. This is why PCR instruments use modules rated for cycling, mounted so that the layers can flex, and why the module, like the fan beside it, is treated as a part with a service life.
Peltier modules are often praised as having nothing to wear out. They have no moving parts, but thermal cycling is their harshest duty, and their life is finite and measured in cycles. The fan that cools the heat sink is a moving part. The modules and the fan together set the instrument's service interval.
The block, the lid and the sensor
The block's job is to make every well the same temperature at the same moment. Aluminum is common. Perkin-Elmer, an early maker of cyclers, filed a patent in 1992 that describes a block machined from a solid piece of a corrosion-resistant aluminum alloy, chosen for thermal uniformity. Silver conducts heat almost twice as well as pure aluminum: 429 watts per meter per kelvin against 237. Some instruments use silver or gold-plated silver blocks.
The speed of a change in temperature, in degrees per second, is called the ramp rate. One manufacturer lists the maximum ramp rate of one of its instruments at 2.6 °C per second with a silver block and 2.3 with an aluminum one, so nearly twice the conductivity gives about 13 % more speed. Conductivity is only one limit, beside the power of the modules and the heat the block must store. Some designs spread heat with a sealed chamber in which a fluid evaporates at the hot side and condenses at the cold, the principle of a heat pipe.
The heated lid sits on the tube caps or the plate seal at about 104 to 105 °C. Without it, water would evaporate from the hot liquid and condense on the cooler cap, changing the volume and concentration of the reaction a little more in every cycle. The lid also presses the tubes down into the block, because a tube that sits loosely in its hole has an air gap around it and heats more slowly than its neighbors.
The temperature sensor is in the block, not in the liquid. The design described in the patent mentioned above has an electronic sensor in a hole drilled into the block, next to a column of wells. The controller reads the sensor many times a second and adjusts the current in the modules so that the block follows the programmed profile. The liquid sits behind a plastic wall and runs a few seconds behind the block, and Chapter 10 shows that handling this lag is the central problem of temperature control.
Not every block instrument uses Peltier modules. The design in the same early patent heated its block with a thin electrical heater and cooled it with chilled liquid from a refrigeration unit, pumped through channels in the block. The Peltier module was adopted in most later designs for its compactness and its ability to both heat and cool, not because nothing else works.
Other ways to change temperature
Changing the temperature of a metal block is one solution, but every change must heat or cool the block's mass along with the sample. Other architectures avoid the block.
In 1989 Carl Wittwer and colleagues described a cycler that put the reaction in thin glass capillaries and heated them with a stream of hot air, venting room air in to cool them. Glass capillaries hold little liquid and have a large surface for their volume, and air has little mass to heat, so the sample changes temperature quickly. Roche's original LightCycler instruments used this design. For the later of them, Roche specified heating a 20-microliter capillary from 40 to 95 °C in under 15 seconds. Another design spins its tubes in a rotor inside a chamber of moving air. Because every tube passes through the same air, its maker claims a temperature uniformity of ±0.02 °C, measured as a standard deviation.
Microfluidic designs can move the sample instead of changing the temperature. In a 1998 demonstration, a glass chip carried the reaction mixture along a channel that wound back and forth across three zones held at fixed temperatures, so that each pass through the zones was one cycle. Twenty cycles took between 90 seconds and about 19 minutes, depending on the flow rate.
Other demonstrations have let natural convection carry the liquid between a hot bottom and a cooler top. Another heated a thin gold film under the wells with blue light and completed 30 cycles in 5 minutes. The authors of a review of microfluidic PCR sort these designs into two groups: in one, the sample stays put and the temperature cycles, so the profile can be changed in software, and in the other, the sample moves through fixed zones, so the cycle is built into the channel.
For anyone specifying, buying or servicing an instrument, three things follow from this chapter. The Peltier modules and the fan have finite lives, so the service plan should state when they are checked or replaced. The block, the lid and the tube or plate form one thermal system, so a change of plate type or seal changes the sample's temperature. And the controller knows only the block's temperature, so the instrument's real performance must be measured in the wells, a subject Chapter 10 takes up.
- Heat flows from the liquid through a plastic wall, a metal block, thermoelectric modules and a heat sink to the room and back, and the resistance and stored heat of each layer shape how the instrument behaves.
- Peltier modules heat or cool by the direction of their current, with resistive losses that make cooling slower, and they wear out under thermal cycling, failing in tests after tens of thousands of cycles.
- The block spreads heat, the lid at about 105 °C stops condensation and presses tubes into contact, and the sensor sits in the block, not in the sample.
- Air-heated capillaries, rotors, flow-through chips, convection and light are other ways to cycle, and they avoid the block's mass at the price of different limits.
Fast, even and exact.
+ The questionWhy is changing a block's temperature quickly and evenly, without overshoot, the hard part of the engineering?
The sample is behind the block
The controller measures the block. The chemistry happens in the liquid. Between them is a plastic wall a fraction of a millimeter thick and the liquid's own bulk, and heat takes time to cross both. When the block jumps to a new temperature, the liquid follows it smoothly, covering a fixed fraction of the remaining difference in each interval of time. Engineers describe this as a first-order lag and characterize it by a time constant: the time the liquid takes to cover about 63 % of a step. In Perkin-Elmer's 1992 patent (Chapter 9), the inventors measured a time constant of about 9 seconds for 100 microliters in a standard tube.
In this example the block jumps from 55 to 95 °C, a step of 40 degrees, and the instrument holds it there for 3 seconds, starting the clock when the block arrives. With a time constant of 9 seconds, the liquid covers 63 % of a step in 9 seconds, leaving 37 %, and 95 % in 27 seconds. It covers the same share of the remaining gap in every 3 seconds, and 0.72 × 0.72 × 0.72 is about 0.37, so each 3 seconds leaves about 72 % of the gap and covers about 28 %. In the 3-second hold it therefore covers 28 % of 40 degrees, about 11 degrees, and reaches only about 55 + 11 = 66 °C before the block is cooled again. At that temperature no target strand would separate. The 9-second figure is for 100 microliters, and a 20-microliter reaction in a thin-walled plate, like the CDC test's, responds faster. The principle is the same: a 3-second denaturation is meaningful only if the liquid itself reaches 95 °C.
Two engineering responses solve the problem of the lag, and modern instruments use both. The first is to calculate the sample's temperature instead of measuring it. The controller knows the block's temperature history, the volume of liquid the user has entered and the time constant of the tube, and from these it computes, moment by moment, where the liquid's temperature must be. That calculation is the sample-temperature model. The hold timer starts when the calculated sample temperature reaches the set point, not when the block does.
The second response is to overshoot. To make the liquid arrive sooner, the controller drives the block deliberately past the set point for a moment and then brings it back, so that the liquid approaches its target quickly without passing it. The early patent describes exactly this: a model of the sample with a single time constant, a block driven past its set point, and a control loop running on the instrument's processor. The loop is of the proportional-integral-derivative kind used throughout industry.
Current instruments leave the choice to the user. One manufacturer's cyclers offer a "calculated" mode, which uses the sample-temperature model, and a "block" mode, which simply holds the block. Sample-temperature control of this kind is what makes a 3-second hold meaningful. Such control also makes the reaction volume entered in the software part of the temperature control: if a user enters 50 microliters for a 20-microliter reaction, the model is wrong.
How fast is fast
Speed is quoted as a ramp rate, in degrees per second (Chapter 9). The number on a datasheet is easy to misread, because manufacturers define it differently: some quote the maximum rate the block reaches for a moment, some the average rate of the block over a temperature change, and some the average rate of the calculated sample.
For the instrument named in the CDC test, the manufacturer's brochure gives a peak block heating rate of 5.5 °C per second. The same brochure gives a sample ramp rate of 3.5 °C per second in the instrument's fast mode and 1.6 °C per second in its standard mode. Another manufacturer gives a maximum of 5 °C per second and an average of 3.3, without saying whether either refers to the block or the sample. A third gives a maximum block rate of 6.5 and an average sample rate of 3.66.
These numbers are not comparable unless their definitions match, and even then a maximum rate may be held only briefly. One instrument maker makes the practical point in its own educational article: the honest test of speed is to run the same protocol on two instruments and time it. For the user, what matters is the time the liquid spends getting from one temperature to the next, added up over every rise and fall in the run: two in each cycle of a two-step protocol such as the CDC test's, and three in a three-step one.
The ramp rate on a datasheet usually describes the block, often at its fastest moment. The liquid follows more slowly, and the difference grows with reaction volume and tube wall thickness. A fair comparison states which rate is quoted, over which temperature range and with which plate and volume, and it times runs under one protocol instead of comparing headline numbers.
Every well the same
A 96-well block is a slab of metal several centimeters across, and its edges lose heat to the surrounding air and to the parts that hold it. Left alone, the corner and edge wells would run cooler than the center when the block is hot and warmer when it is cold. Designers counter the edge loss with geometry and with extra heat. The design in the early patent had a groove cut around the perimeter of the block, a guard band that limited heat flow from the central wells to the edge. It also had separately controlled heaters along the edges to replace the heat lost there.
The result is specified as temperature uniformity, which says how far apart the wells are from one another. It is a different property from accuracy, which says how close the average well is to the set point: a block can be perfectly uniform and two degrees off, or accurate on average with a cold corner. For uniformity, too, the definitions on datasheets differ. One manufacturer states ±0.4 °C well to well, measured within 10 seconds of the block arriving at 90 °C. For its later model it states ±0.3 °C, with the hottest and coldest wells no more than 0.6 °C apart, measured 10 seconds after the block reaches its target. Between the two models the same maker changed the stated moment of measurement and added a limit on the hottest-to-coldest spread, so even its own two figures do not share one definition. The brochure for the instrument named in the CDC test states a uniformity of ±1 °C and an accuracy of ±0.5 °C without stating the conditions.
Some instruments make the block uneven on purpose. A gradient block deliberately holds a range of temperatures across the block, so that a developer can test several annealing temperatures in one run, and in one design the range can be up to 24 °C wide. Another design divides its block into six independently controlled zones that can differ by up to 5 °C.
When temperature goes wrong
Temperature errors are not hypothetical. In a study published in 2005 in a clinical microbiology journal, researchers tested six new 96-well Peltier cyclers. They measured thirteen positions in each block with a measuring system certified to ±0.3 °C, and then ran a diagnostic PCR on each cycler. The errors were largest in the first 15 seconds of the denaturing step. When the hold timer started, the liquid in the accurate instruments was 0.5 to 1.5 °C from the set point, while in two less accurate instruments it was still 13 to 20 °C short. With those two the diagnostic test gave no positive results until the denaturation temperature and time were increased. The authors named the critical parameters: holds timed too early, undershoot and overshoot, and differences in heat transfer across the block.
A cool well at the annealing step lets primers bind to near-matches, which can produce wrong products (Chapter 5). A cool well at the denaturing step leaves strands paired, lowering the efficiency, and every cycle of lower efficiency delays the curve (Chapter 6). A hot well uses up the enzyme's life faster (Chapter 7). The effects compound over forty cycles, so a well that is consistently a degree or two off can give a result that differs from its neighbors', and a well that is grossly off can give no result.
Measuring what the wells see
Since the controller reads only the block, the wells must be measured separately. Temperature verification kits place calibrated probes in selected wells and record what they read through a test program. One manufacturer measured eleven cycler models from six manufacturers with such a kit, whose probes were calibrated traceably to national standards. It found steady-state errors between −0.33 and +0.39 °C, and measured performance that did not always match the published specifications. The manufacturer that ran the study competes with the makers of most of the instruments tested, so the study is best read as an illustration of method, not as a ranking.
Calibration services offer the same measurements: steady accuracy, well-to-well spread, overshoot and undershoot, and stability during holds. Accredited services work under ISO/IEC 17025, the international standard for the competence of testing and calibration laboratories. One such service states an overall accuracy better than 0.1 °C and recommends calibration once a year or every 250 runs. CYCLERtest sells cycler calibration systems, and in a metrology paper presented in 2017 two of its staff and a coauthor argued that methods which judge a cycler by running a PCR on it cannot be traced to physical units and carry large uncertainties. They also argued that acceptance limits should be derived from the needs of the user's own PCR method, not taken from the manufacturer's specification. The point generalizes: how much error a cycler may have is a property of the test it runs.
When specifying or buying a cycler, ask for accuracy and uniformity with their conditions: at which temperature, how long after arrival and with which plate. Ask for ramp rates with their definitions, and for the assumptions of the sample-temperature model. At installation and at intervals afterward, measure the wells with traceable probes and compare the results with limits derived from the tests the instrument will run.
- The liquid lags the block with a time constant of seconds, so instruments calculate the sample's temperature, overshoot the block deliberately and start hold timers when the sample arrives.
- Ramp rates are defined in several incompatible ways, and the speed that matters is how fast the liquid moves between temperatures, best compared by timing the same protocol.
- Edge wells lose heat, so blocks use guard bands and edge heaters, and uniformity and accuracy are distinct specifications, each with conditions attached.
- Real instruments can start a hold before the liquid arrives: in one study, the liquid in two of six new cyclers was still 13 to 20 °C below the denaturing temperature when the timer started, so wells must be measured with traceable probes against limits set by the test.
+ Part IV · Counting
From copies to a number.
In 1993 Russell Higuchi and colleagues at Roche used a video camera to watch many reactions at once, cycle by cycle, and saw that the fewer cycles a reaction needed to reach detectable fluorescence, the more target it had started with. Copying answers whether a target is there, but the central question also asks how much. This part keeps the chemistry fixed and changes only when and how the product is read: once at the end, in every cycle, or one partition at a time. It shows how light reports the copies, how the cycle at which the light crosses a line becomes a quantity, what that number can and cannot be trusted for, and how splitting a sample into thousands of partitions counts molecules directly. By the end of this part the central question is answered.
Endpoint, real-time and digital.
+ The questionOnce PCR could copy, what did each generation add to answering "how much"?
One chemistry, three ways to read it
The cycle described in Part II has not changed in its essentials since Taq polymerase made it automatic in 1988: denature, anneal, extend, repeat. What has changed is when and how the product is observed, and with it the questions PCR can answer. The history is best told by holding the chemistry constant and asking, for each generation, what it reads, when it reads it, and therefore what question it can answer. Each answers a sharper question than the last.
Endpoint: is it there, and is it the right size?
The first generation ran the reaction to completion and then looked at the product, which is why it is called endpoint PCR. The standard way to look was gel electrophoresis. DNA's backbone carries a negative charge on every phosphate group, so in an electric field DNA moves toward the positive electrode. The gel is a slab of agarose, a seaweed polysaccharide that forms a mesh of pores, and short pieces of DNA move through the mesh faster than long ones. After half an hour or so, the products of each reaction have spread out by size. A lane of DNA fragments of known lengths, run alongside, acts as a ruler, and a fluorescent stain makes the DNA visible under ultraviolet or blue light.
A band at the expected position says that the reaction made a product of the expected size, which is good evidence that the primers found their target. According to a stain supplier, a band stained with ethidium bromide needs roughly half a nanogram to five nanograms of DNA, which for a 200-base-pair product is roughly 2 billion to 20 billion copies (Chapter 1). More sensitive stains need less, but a band still becomes visible only well into the reaction, near or at the plateau. The endpoint therefore answers the question "is it there?" well and the question "how much?" badly, because reactions started from very different amounts end at similar plateaus (Chapter 8).
Endpoint detection has two other weaknesses. An empty lane can mean that the target was below the limit or that the reaction failed, and a positive control run alongside can tell the two apart only for the run as a whole (Chapter 8). It shows that the reagents and the instrument worked, but a failure in one sample's own reaction, from a substance in the specimen that slows the enzyme, for example, shows only in a control amplified from that same sample (Chapter 21). Opening the tube after amplification also releases billions of copies of product into the laboratory, where a trace can reach the next reaction and give a false positive. Chapter 20 describes the chemical defense against such carry-over.
Real-time: how much, compared with known amounts?
The second generation kept the tube closed and watched the reaction as it ran. Russell Higuchi and his colleagues worked in the PCR group that passed from Cetus to Roche in 1991. In 1992 they added ethidium bromide to the PCR itself. This dye fluoresces much more strongly when it is bound to double-stranded DNA, and they followed its fluorescence through the wall of the tube as the product accumulated. In 1993 they used a video camera to watch many reactions at once, cycle by cycle, and observed the relationship that real-time PCR has used ever since: the fewer cycles a reaction needs to reach detectable fluorescence, the more target it started with.
Reading in every cycle changes what can be measured. At the plateau, reactions that started from different amounts look alike, but in the exponential phase they stay apart: a tenfold difference in starting target appears as a steady gap of about 3.3 cycles (Chapter 6). The instrument records the cycle at which each reaction's fluorescence crosses a fixed line. Recording that cycle turns the gap into a number, and samples of known amount, run alongside, allow the number to be converted into a quantity. Real-time PCR answers "how much, relative to known standards?" In 1996 researchers at Genentech and at the Applied Biosystems division of Perkin-Elmer, which made the instrument they used, described real-time PCR with fluorescent probes. They reported a working range of at least five tenfold steps, or 100,000-fold, with no handling of the product after amplification. Chapters 12 and 13 show how the light is produced and how the number is read.
Two reactions, chosen for illustration, start with 1,000 and 1,000,000 copies of the target, and both are assumed to amplify with the same, near-perfect efficiency. The second holds 1,000 times as much, which is three tenfold steps, because 10 × 10 × 10 = 1,000. At about 3.32 cycles per tenfold step, its curve crosses the line about 3 × 3.32 ≈ 10 cycles earlier, so the real-time read reports the difference as a gap of about 10 cycles. By the last cycle both reactions have reached similar plateaus, so their bands on a gel look alike, and the gel reports only that a product of the right size is present in both. Across the working range reported in 1996, at least 100,000-fold, a difference in starting amount shows as a difference in crossing cycle.
Digital: exactly how many molecules?
The third generation abandoned the curve altogether. A sample can be diluted and split into many separate reactions, or partitions, so small that most contain one target molecule or none. Each little reaction then either amplifies or does not, and counting the reactions that light up counts the molecules. In a 1992 study, researchers used this limiting-dilution principle, with statistics, to count the targets of a leukemic cell clone, and they detected about two such targets among 160,000 normal genomes. In 1999 Bert Vogelstein and Kenneth Kinzler named the approach digital PCR. They described it as turning the exponential, analog output of PCR into a linear, digital count, and they used it to detect a cancer-causing mutation in DNA shed into the stool of patients with colorectal cancer.
Digital PCR answers "exactly how many target molecules were in this volume?" without a curve and without standards of known amount. It does depend on knowing the volume of each partition, and on a statistical correction for partitions that receive more than one molecule (Chapter 14).
Each generation matches a different need. An endpoint reaction with a gel is still the cheapest way to check that a product has the expected size, and endpoint PCR prepares DNA for sequencing and cloning. Real-time PCR is the form of PCR most used in diagnostics, because it is closed, fast and quantitative enough. Digital PCR is used where an absolute count or a very rare target matters, such as a tumor mutation present in a small fraction of the DNA in a blood sample.
The evolution at a glance
| Generation | When it reads | What it reads | The question it answers |
|---|---|---|---|
| Endpoint | once, after the last cycle | product size and presence, on a gel | Is it there, and is it the right size? |
| Real-time | in every cycle | fluorescence rising with the product | How much, compared with known amounts? |
| Digital | once, after cycling, partition by partition | the number of positive partitions | Exactly how many molecules? |
The chemistry in all three is the same primers, polymerase and cycle. What changed is when the light is read and how many separate reactions the sample is divided into.
For a test developer, the first design question is which of the three questions the result must answer. A yes-or-no infectious-disease test needs a reliable call at a cut-off, which real-time PCR provides. A viral-load test needs a quantity, with standards and an efficiency that holds across the range. A test for a rare mutation needs to count a few mutant molecules among many normal ones, which is where partitioning justifies its cost.
- The chemistry has stayed the same, and the generations differ in when and how the product is read.
- Endpoint PCR reads the product on a gel after cycling: good for presence and size, poor for quantity, and it opens the tube to contamination.
- Real-time PCR, from 1992 to 1993, reads fluorescence in every cycle in a closed tube, and the cycle at which the signal rises reveals the starting amount relative to standards.
- Digital PCR, named in 1999, splits the sample into many partitions and counts the positive ones to give an absolute number of molecules.
Seeing what can't be weighed.
+ The questionHow does an instrument watch copies form inside a sealed tube, and tell several targets apart?
Light in, different light out
A real-time instrument cannot weigh the product or look at it under a microscope. It can only shine light into the well and measure the light that comes back. The method that makes such a measurement possible is fluorescence. Some molecules, called fluorophores, absorb light of one color and give off light of a longer wavelength, a slightly redder color, a few billionths of a second later.
The instrument shines light of the absorbed color, the excitation, through a filter that passes only that band, and measures the returning light through a second filter that passes only the emitted band. Such a pair of filters, one for the light going in and one for the light coming out, makes one color channel. Because the two bands differ, the detector sees the fluorophore's glow and very little of the much brighter excitation light. What the instrument measures, then, is not DNA but light from a dye, and the task of real-time detection chemistry is to make that light rise in proportion to the copies of the target. Two families of chemistry do this.
Dyes that light up on any double-stranded DNA
The simplest approach adds a dye whose fluorescence increases strongly when it binds double-stranded DNA. Higuchi's group at Roche used ethidium bromide (Chapter 11). Most laboratories now use dyes such as SYBR Green I. As product accumulates, there is more double-stranded DNA for the dye to bind, and the well glows more brightly. SYBR Green I is often called an intercalating dye, one that slides between base pairs, but a study of how it binds found that it intercalates only at low ratios of dye to DNA and gives most of its signal by binding the surface of the helix.
The weakness of such a dye is that it reports any double-stranded DNA, including primer-dimers and other wrong products. The standard check is a melt curve. After the last cycle, the instrument heats the product slowly while watching the fluorescence. As each kind of product reaches its melting temperature (Chapter 2), its strands separate and the dye lets go, so the fluorescence drops. Plotted against temperature, the rate of that drop shows a peak at the melting temperature of each product: one sharp peak suggests one product, and a second, lower peak often reveals primer-dimers. One supplier states in a technical note that a single product with regions of different G–C content can give two peaks, so a melt curve is an indicator, not proof, and a gel or sequencing confirms the product.
Probes that light up only on the target
Diagnostic tests mostly use a probe, a third short piece of synthetic DNA that binds inside the target between the two primers. Like the primers, it is an oligonucleotide: a short strand of DNA made to order. The probe carries a fluorophore at one end and a quencher at the other. A quencher is a molecule that absorbs the energy of a fluorophore close to it. While the probe is intact, its two ends are close enough for the quencher to absorb that energy, and the probe is dark.
The most widely used design relies on the activity of Taq described in Chapter 7. As the polymerase extends a primer along the template, it meets the probe bound in its path and cuts the probe apart with its 5′ to 3′ exonuclease activity. The fluorophore, now separated from the quencher, glows. In each cycle the polymerase cuts probes on every newly copied target, so the fluorescence rises with the product, but only with the product the probe recognizes.
Scientists at Cetus showed the method with a cleavable probe in 1991. By 1993 scientists at Perkin-Elmer's Applied Biosystems division had designed probes that carried a fluorescent reporter and a quencher, and in 1996 researchers at Genentech and at the same division, which made the instrument they used, described real-time quantitative PCR with these probes (Chapter 11). Such a probe is often called a TaqMan probe or a hydrolysis probe. Other designs work differently, and molecular beacons, introduced in 1996, are one example: hairpins that open and glow when they bind a matching target.
A probe must bind as well as both primers, so a probe test has a third check on identity, and that check makes it far less likely than a DNA-binding dye to report primer-dimers or wrong products. Even so, a poorly designed probe can still produce signal without target (Chapter 22).
In its form authorized from March 2020, the CDC test used one hydrolysis probe for each of its two viral targets and one for a human control gene, and each carried the fluorophore FAM at one end and a quencher called Black Hole Quencher 1 at the other. All three probes used the same fluorophore, so each specimen was run in three separate wells, one per target. The instrument read the fluorescence at the 55 °C step of each cycle, when the probes are being cut. In March 2020 the FDA also accepted probes with an additional internal quencher. The test's instructions told users to set the quencher to "none" in the instrument's software, but that setting is for a quencher that emits no light of its own, not for a probe without a quencher.
The hardware of a reading
An optical system needs a light source, filters and a detector, and real instruments combine them in several ways. The instrument named in the CDC test uses a 75-watt halogen lamp, a set of five excitation filters and five emission filters, and a CCD camera that images the whole plate at once. Another instrument uses a 100-watt xenon lamp and a cooled camera.
Newer designs use light-emitting diodes, or LEDs. One uses a white LED and a CMOS camera with six excitation and six emission filters. Another uses six colored LEDs and six filtered photodiodes, mounted on a small carriage that moves over the plate and reads each well in turn, so that every well is read through the same light path. According to its maker, this carriage takes about 12 seconds to read all channels of a 96-well plate. A rotor instrument reads each tube as it spins past a photomultiplier tube, a very sensitive detector.
Each architecture has a cost. A camera reads all wells at once but views each well from a slightly different position and angle, so it needs calibrations that locate every well on the image and correct for each well's position. A scanning carriage reads every well identically but takes time, which must fit inside the cycle step during which the reading is valid. Lamps give broad, bright light, but they age. LEDs are stable and long-lived, and each colored LED gives one narrow band.
Every number from a real-time instrument begins as light from a dye, measured through a filter and corrected by calibrations. A dirty filter, a fading lamp or a bubble in the well changes the light without changing the copies, and so does a wrong pure-dye calibration, which the next section describes. That is why the calibrations are scheduled, why instruments are verified with a reference plate, and why Chapter 15 traces every conversion between the copies and the reported result.
Several targets in one well
An instrument with five or six channels can, in principle, read five or six fluorophores in the same well, one in each channel. One instrument shows the pattern. Its first channel excites at 450 to 490 nanometers and reads 510 to 530, for FAM. Its second excites at 515 to 535 and reads 560 to 580, for a dye called HEX, and the others continue in the same way up to the near infrared. The CDC's later multiplex test detected four targets in one well, influenza A, influenza B, SARS-CoV-2 and the human control, with four different fluorophores.
In the form authorized from March 2020, the CDC test detected three targets in each specimen, and all three probes carried FAM (Chapter 22). Three signals of the same color in one well could not be told apart, so each specimen took three wells, and a 96-well plate held at most 96 ÷ 3 = 32 specimens before any wells went to controls. The CDC's later multiplex test detected four targets with four different fluorophores, so each specimen took one well. On a 96-well plate that is up to 96 specimens before controls, three times as many, and each well gives four readings instead of one. The price is that four dyes share one well, so the instrument needs a channel for each of them and must separate their light, as the next paragraphs describe.
Multiplexing has a cost: fluorophores emit over broad bands, so a bright dye leaks some light into a neighboring channel. Instruments correct for this leak with calibration. The maintenance guide for one family of instruments lists the calibrations its owner must run. A region-of-interest calibration maps where each well falls on the camera image, every six months or after a lamp change. A background calibration measures the instrument's own fluorescence, monthly. A pure-dye calibration records each dye's spectrum on that instrument, every six months. During a run the software uses the pure-dye spectra to work out how much of the measured light came from each dye, a step called multicomponent analysis.
Many master mixes also contain an inert reference dye, ROX, which serves as a passive reference. Some instruments divide the reporter signal by it, and the division corrects for differences in volume, bubbles and condensation. The CDC test's instructions had users switch that correction off by setting the passive reference to "none", although its original master mix contained ROX.
Choosing fluorophores for a test is an optical design decision. Each dye must sit in a channel the instrument has, be calibrated on that instrument, and be far enough in wavelength from its neighbors that the unmixing is reliable at the brightest signal the test will produce. The instrument's list of calibrated dyes is therefore an input to assay design, not an afterthought.
- Real-time instruments measure fluorescence: light of one color in, longer-wavelength light out, separated by filters.
- DNA-binding dyes glow with any double-stranded DNA and need a melt curve as a check, while hydrolysis probes glow only when the polymerase cuts them on the intended target.
- Lamps or LEDs, filters and cameras, photodiodes or photomultipliers combine in different architectures, each with its own corrections.
- Several dyes in one well need distinct channels, pure-dye calibration and unmixing, and the instrument always reads light, never the copies directly.
What a Ct value means.
+ The questionWhat does a Ct value measure, and how far can it be trusted as a quantity?
From light to a curve
For each target that amplifies in a well, a real-time run reports one number, such as 27.4. That number begins as light. In every cycle, at the step the test specifies, the instrument records the fluorescence of every well in every color channel (Chapter 12). Plotted against cycle number, the readings for a positive well trace the S shape of Chapter 8: flat for many cycles, then a steep rise, then a plateau. The flat early part is not zero. It is the baseline, the background light from the dye, the plastic and the optics that is present before any product accumulates. The analysis software estimates the baseline from early cycles, typically somewhere between cycles 3 and 15, and subtracts it, so that every curve starts from a common floor.
Next comes a horizontal line, the threshold, placed above the background noise and well below the plateau, in the region where the curves are rising steadily. The cycle at which a well's curve crosses the threshold is the number the instrument reports. The software interpolates between readings, so the number can be a fraction such as 27.4. Manufacturers named it differently: Applied Biosystems called it the threshold cycle, Ct, and Roche called it the crossing point, Cp. The MIQE guidelines of 2009 proposed a neutral name, the quantification cycle, or Cq. This guide uses Cq alongside the familiar Ct.
The CDC's instructions for its test show how much of this is in the user's hands. They instructed laboratories to leave the baseline at the software's default and to set the threshold by hand, dragging it into the exponential part of the curves, above the background. The threshold is a setting, and the Ct depends on it.
Why the crossing cycle measures the start
The threshold corresponds to a fixed amount of product, so every well that crosses it holds roughly the same number of copies at that moment. A well that started with more target gets there in fewer doublings. This is the backward reading of the fact that a cycle can at most double the target. If one sample starts with ten times more target than another, it reaches any given amount of product earlier, by as many cycles as it takes to multiply by ten, and at perfect doubling that is 3.32 cycles, because 3.32 doublings make a factor of ten.
Two samples, chosen for illustration, cross the threshold at Cq 18 and Cq 28, a gap of 10 cycles. At 100 % efficiency each cycle doubles the target, so the sample at Cq 18 started with 2¹⁰ = 1,024 times as much. At 90 % efficiency each cycle multiplies the target by 1.9, and the same gap means 1.9¹⁰, or about 613 times as much. At perfect doubling, 1 cycle means 2-fold, 3.32 cycles 10-fold, 6.64 cycles 100-fold and 9.97 cycles 1,000-fold. A practitioner carries the result as a rule of thumb: a lower Cq means more target, and about 3.3 cycles is a tenfold change.
The arithmetic rests on an assumption that is easy to forget: the efficiency is known, and it is the same in the samples being compared. Two samples may start with the same amount of target and still amplify with different efficiencies, for example because one carries an inhibitor from the specimen, a substance that lowers the efficiency of amplification. The two then cross at different cycles, and the difference between them is misread as a difference in amount.
The standard curve
To turn Cq into a number of copies, a quantitative test runs standards: a dilution series of known amounts of target, for example in tenfold steps from ten copies to a million. When each standard's Cq is plotted against its amount on a scale of tenfold steps, the points fall on a straight line, the standard curve, and an unknown sample's Cq can then be read off the line as an amount.
The slope of the line reports the efficiency. Perfect doubling gives a slope of −3.32 cycles per tenfold step, negative because each tenfold increase in the starting amount lowers the Cq by 3.32 cycles. At 90 % efficiency the slope is −3.59, because each tenfold increase takes more cycles. A slope of −3.10 corresponds to an apparent efficiency of 110 %, which would mean more than doubling in every cycle. That is physically impossible, so such a slope signals that something other than clean amplification is shaping the curves.
A range of 90 to 110 % is widely used as an acceptance criterion for efficiency, together with a straight-line fit whose R² is at least 0.98. R² measures how closely the points follow the line, and a value of 1 means that they lie on it exactly. The 90 to 110 % range and the R² of 0.98 come from laboratory practice, for example the criteria of the European network of laboratories that test for genetically modified food, not from the MIQE guidelines, which set no fixed range.
The 2025 revision of MIQE asks for more than a slope, and it recommends three things. Cq values should be converted into quantities corrected for efficiency and reported with intervals that show their uncertainty. The raw fluorescence data should be exported in vendor-neutral formats, not tied to one maker's software. And each assay should state both its limit of detection and its lower limit of quantification, the smallest amount it can measure with useful precision. The revision notes that the limit of detection cannot be below about three copies (Chapter 6) and that a good assay quantifies from about ten.
Comparing samples without standards
Many research studies ask whether a gene's activity changes, not how many copies there are. A widely used shortcut for them is the ΔΔCt method, which Kenneth Livak and Thomas Schmittgen published in 2001. Livak worked at Applied Biosystems, which made real-time instruments, and Schmittgen at Washington State University. The method first compares each sample's target with a reference gene in the same sample, a gene whose activity is assumed to stay constant. It then compares those differences between samples and converts the result into a fold change with powers of two. The method works only if the target and the reference amplify with nearly the same, nearly perfect efficiency. The 2025 MIQE revision strongly recommends against relying on it and favors efficiency-corrected quantities instead.
Reading a qualitative result
Most infectious-disease tests do not report a quantity at all. They compare each Cq with a cut-off and apply rules that combine the targets and the controls. In its form authorized from March 2020, the CDC test used three wells for each specimen: two for regions of the virus's genome, named N1 and N2, and one for RNase P, a human gene (Chapter 12). It counted a target as detected if its curve crossed the threshold before cycle 40.00 of its 45, and it read each specimen's wells together with the run's controls. These were a no-template control, a positive control and a human specimen control, a specimen-like sample carried through extraction (Chapter 21). The table applies these rules to a set of wells. Its Cq values are illustrative, chosen to show each rule, and do not come from a real run.
| Well | N1 | N2 | RNase P | Reading under the CDC rules |
|---|---|---|---|---|
| No-template control | none | none | none | Valid: no amplification anywhere, so no contamination detected |
| Positive control | 26.1 | 27.4 | 28.0 | Valid: all three targets detected before cycle 40 |
| Human specimen control | none | none | 27.0 | Valid: extraction worked and carried no viral target |
| Specimen A | 24.3 | 25.9 | 26.8 | Positive: both viral targets detected |
| Specimen B | none | none | 25.5 | Negative: no viral target, and the human control shows the specimen was adequate |
| Specimen C | 38.6 | none | 27.2 | Inconclusive: only one viral target; repeat the test |
| Specimen D | none | none | none | Invalid: not even the human control; re-extract and retest, and if still invalid, consider a new specimen |
| Specimen E | 18.2 | 19.0 | none | Positive: both viral targets; a missing human control does not invalidate a positive |
The last row can look like an error. The human control gene, RNase P, is there to show that a negative result is believable: that the swab collected human cells, that extraction recovered their nucleic acid and that amplification worked in the control's own well (Chapter 21). A specimen in which both viral targets are detected has already shown that amplification worked, so under the CDC rules its positive result stands whether or not the control is detected.
A Ct in this table, such as Specimen E's 18.2, does not measure how much virus the patient carries. A qualitative test runs no standards, so its Ct cannot be converted to a quantity, and the value moves with the specimen, the extraction, the instrument and the threshold. Only a validated quantitative test, with standards and a known efficiency, turns a Cq into a quantity.
What a Ct cannot show
A Cq is a property of one test, on one instrument, with one threshold. The same sample can give very different values on different tests. Canada's public health agency reports that identical samples tested in several provinces gave Ct values up to 8 cycles apart, for example 22 on one test and 30 on another, a gap that would mean a difference of more than a hundredfold if it were read naively as a difference in amount. A statement by the American Association for Clinical Chemistry notes that there is no international unit for Ct. A joint statement by two professional societies lists the reasons the values differ: the collection device, the specimen type, the extraction, the genetic target and the chemistry all shift the Ct.
During the COVID-19 pandemic, many people hoped Ct would show whether a patient was infectious. Studies that tried to grow live virus from positive samples illustrate the difficulty. A 2020 Canadian study cultured 90 samples and grew no virus from any with a Ct above 24, while a 2020 English study of 324 samples found that the chance of growing virus fell to 8 % at Ct values above 35. The two studies used different tests, culture methods and laboratories, and the highest Ct at which each still grew virus differs by more than ten cycles. That gap cannot be assigned to the tests alone, but it shows that a Ct cut-off found in one setting cannot be carried to another. The professional bodies advised laboratories not to report Ct for routine patient management, or to report it only with an explanatory comment.
A qualitative test was validated to give a positive, negative, inconclusive or invalid result at its cut-off, and that is what it should report. A quantitative test reports a quantity with its uncertainty and its limits of detection and quantification, and it keeps its raw data in a form that can be re-analyzed. Software that shows Ct values to the users of a qualitative test invites the misreading this chapter describes.
- The software subtracts a baseline, a threshold is set in the exponential region, by the software or by hand, and the instrument reports the fractional cycle at which each curve crosses it: Cq, often called Ct.
- Because each cycle at most doubles the target, a tenfold difference in starting amount shifts Cq by about 3.3 cycles, if efficiencies are equal and near perfect.
- Standard curves convert Cq to a quantity and reveal the efficiency, and the 2025 MIQE revision asks for efficiency-corrected quantities with their uncertainty and recommends against relying on the ΔΔCt shortcut.
- Qualitative tests compare Cq with a cut-off under rules that include controls, a Ct from such a test is not a viral load, and no Ct is comparable across tests.
Counting without a curve.
+ The questionHow can splitting one sample into thousands of partitions count molecules without a standard curve?
Split, amplify, count
Real-time PCR infers the starting amount from how long the reaction takes to become visible, and that inference depends on efficiency and standards. Digital PCR asks a simpler question of many small reactions instead of a hard question of one large one. The reaction mix, with the sample in it, is divided into thousands of tiny partitions (Chapter 8). They are droplets of liquid separated by oil, or chambers etched in a chip, and each holds a fraction of a nanoliter. The whole set is cycled. A partition that received at least one target molecule amplifies it to a bright signal, while a partition that received none stays dark. The instrument then counts the bright partitions.
The count does not depend on efficiency in the way a Cq does (Chapter 13). A partition that amplifies a little slowly still ends up bright after forty cycles, and it is counted as positive just the same. And no standards are needed, because the answer is a count of molecules in a known volume, not a comparison with samples of known amount.
Molecules arrive at random
The difficulty is that molecules do not distribute themselves one to a partition. They land at random, so at any useful concentration some partitions receive two or three molecules and still count as one bright partition. Counting bright partitions therefore undercounts the molecules, and the undercount grows with concentration. Digital PCR corrects for the undercount with the Poisson distribution, the statistics of random scattering that set the sampling floor in Chapter 6.
The correction works from the dark partitions, because those are unambiguous: a dark partition received exactly zero molecules. Chapter 6 showed that if the average is one molecule per partition, 37 % of partitions receive none, and each additional molecule in the average multiplies that fraction by 0.37 again. The relationship also works backward, so the fraction of dark partitions gives the average number of molecules per partition.
A run chosen for illustration reads 20,000 partitions, of which 5,000 are bright, so 15,000, or 75 %, are dark. By the rule of Chapter 6, the share of dark partitions is 0.37, more exactly 0.368, raised to the power of the average number of molecules per partition. The average is therefore the power to which 0.368 must be raised to give 0.75, and trying values gives 0.2877, or about 0.288: 0.368 raised to the power 0.288 is 0.75. The partitions therefore held about 0.2877 × 20,000 ≈ 5,750 molecules, about 15 % more than the 5,000 bright partitions. With droplets of 0.85 nanoliters, 0.288 molecules per droplet is 0.288 ÷ 0.85 ≈ 0.34 molecules per nanoliter. A microliter is 1,000 nanoliters, so the reaction held about 340 molecules per microliter. Without the correction, the count would have missed about 5,750 − 5,000 = 750 molecules: the second and further molecules in partitions that held more than one.
The undercount grows quickly. At an average of 0.1 molecules per partition, about 5 % of the bright partitions hold more than one molecule. At an average of one, 42 % do, and at an average of 1.6, about 60 % do. In the popular picture of digital PCR, each partition holds one molecule or none, but that picture is true only at very low concentrations: at the concentrations that give the best precision, most bright partitions hold more than one molecule. The result therefore depends heavily on the correction.
Precision and range
Because the answer comes from counting, its precision follows the statistics of counting. Scientists at a digital PCR manufacturer modeled that precision and found that it is best at an average of about 1.6 molecules per partition, when about 20 % of partitions are dark, and that it improves with the square root of the number of partitions. Four times as many partitions therefore halve the relative uncertainty.
The range has firm limits at both ends. At the top, when nearly every partition is bright, the few dark ones carry all the information: with 20,000 partitions, a single remaining dark partition corresponds to an average of about 9.9 molecules per partition, and beyond that the measurement breaks down. A run with every partition bright, or every partition dark, carries no quantitative information at all. At the bottom, the limit is the volume actually analyzed, because no molecule can be counted that was not loaded. One manufacturer's plates analyze about 3 microliters per well in one format and about 24 microliters in another, and a test that analyzes the larger volume can detect rarer targets.
The partition volume is the calibration
The phrase "absolute quantification without a standard curve" needs one qualification. Digital PCR counts molecules per partition, and converting that into molecules per microliter requires the volume of a partition. If the volume is wrong, every result is wrong by the same factor. In a 2015 study, scientists at the European Commission's Joint Research Centre measured the droplets of a widely used system under a microscope. They found a mean volume of 0.834 nanoliters, about 8 % smaller than the value then built into the instrument's software, and the difference was enough to explain a systematic disagreement with another digital platform. The software value was later changed to 0.85 nanoliters. Digital PCR needs no standard curve, but every partition must have the volume that the software assumes.
Digital PCR's count is absolute only to the accuracy of its partition volume and the uniformity of its partitions. Partitions also differ from one another slightly, and partitions that fall between bright and dark must be classified by a threshold the user or software sets. Each of these is a source of error that a standard curve would have absorbed and that digital PCR must control directly.
The instruments
The systems described here divide a reaction into between about 8,500 and 100,000 partitions, depending on the platform and the format, and they read them in one of two ways. The droplet systems that made the method widespread make about 20,000 droplets from a 20-microliter sample in a separate droplet generator. The droplets are cycled in an ordinary thermal cycler and then streamed single file past a detector in a reader, so the work passes from one instrument to the next, with plates moved between them by hand. QuantaLife developed this system and was sold to Bio-Rad in 2011. That year its scientists, by then at Bio-Rad, published the system's original description and reported droplets of about a nanoliter, generated and read at about a thousand per second. Later droplet readers detect six colors, according to their maker.
Other systems fix the partitions in place and image them. Some integrate everything: the user loads a plate and the instrument partitions, cycles and images it. One offers plates with about 8,500 or 26,000 partitions per well. Another uses 20,480 microchambers per sample and gives results in under 90 minutes, according to its maker. A third partitions plates of 20,000, 28,000 or 100,000 nanowells in a separate step before its analyzer cycles and images them. Another approach packs droplets into a flat single layer in a chip and images it.
The choice between these systems is an engineering trade between throughput, hands-on time, the volume analyzed and the cost per sample. The market is still moving: in July 2025 one of the large suppliers announced it had acquired a smaller company whose droplet-imaging technology it sells as of October 2026, in a series that detects seven colors according to its maker.
Digital PCR justifies its cost where counting matters more than speed. One use is measuring a rare mutation among many normal copies of the same gene, for example a tumor mutation in DNA from a blood sample. Another is measuring copy numbers absolutely, without standards. A third is testing samples that carry substances that slow the reaction, which digital PCR is reported to tolerate better than real-time PCR. Digital PCR is slower and more expensive per sample than real-time PCR and covers a narrower range of concentrations in one run, so it complements real-time PCR rather than replacing it.
The central question, answered
The guide's central question can now be answered in full. Heating and cooling a sealed tube finds one stretch of DNA among billions because at the annealing temperature two primers bind stably only where their sequences match, and only the stretch between two primers facing each other is copied exponentially (Chapters 2 and 5). The same cycle copies that stretch a billion times because each round of separating, priming and extending lets every copy act as a template, so the count nearly doubles in each cycle, and a heat-stable enzyme survives the whole run (Chapters 4, 6 and 7).
The instrument counts how much of the target the sample held because a probe or dye makes light in proportion to the copies, and the cycle at which the light crosses a line, or the fraction of partitions that never light, reports the starting amount (Chapters 12 to 14). The rest of the guide deepens the second half of the answer: how the instrument is engineered so that its temperatures and its readings can be trusted, and how a test built on it earns that trust.
Digital PCR is worth its cost when the result must be an absolute count, when the target is a small fraction of a large background, or when the sample tends to inhibit the reaction. It requires attention to partition volume and uniformity, to the threshold that separates bright from dark, and to loading in the range where the statistics are precise.
- Digital PCR divides the sample into thousands of partitions, cycles them and counts the bright ones, and the count depends far less on efficiency than a Cq does and needs no standards.
- Molecules land at random, so many bright partitions hold more than one molecule, and the fraction of dark partitions gives the true average, about 15 % above the naive count in the chapter's example.
- Precision is best near 1.6 molecules per partition and improves with the square root of the number of partitions, and the volume analyzed sets the lower limit.
- Results are absolute only if the partition volume is right: one study found an 8 % error in a widely used system.
+ Part V · The instrument as a system
Engineering the whole.
The CDC test was authorized for one instrument running one version of its software, and its instructions also fixed the run mode, the dye settings and the way the threshold was set. Those settings were part of the test, because the heater, the optics, the software and the chemistry work as one system. This part looks at the instrument as a whole, the way a systems engineer would. It shows how the parts are orchestrated and how a number is produced, how a stated need becomes a specification for every part, why the instrument and the assay cannot be specified apart, what it takes to integrate a whole laboratory into a cartridge, and how every part and the whole are verified.
The instrument as a system.
+ The questionHow do the heater, the optics and the software act as one instrument, and how do photons become a number?
Subsystems, each with one job
A typical real-time PCR instrument contains a metal block, a light source and a detector, a tray that carries the plate in, and circuit boards. None of these parts knows anything about DNA. The instrument can be described as a set of subsystems, each with one job.
The thermal subsystem delivers the temperature profile to the liquid in every well, using the block, its Peltier modules, the heat sink, the lid and the temperature sensors of Chapters 9 and 10. The optical subsystem measures the fluorescence of every well in every channel, with the light source, filters and detector of Chapter 12. A motion subsystem moves what must move: the tray that carries the plate into the instrument, a filter wheel, or a carriage that scans the optics across the plate.
The electronics supply power and drive the modules with currents of several amperes. They also condition the small signals from the temperature sensors and the light detector and convert them into numbers. Firmware on the instrument's processors runs the control loops and the sequence of events in real time. The software usually runs on an attached computer. It lets the user set up a run, analyzes the data into curves, Cq values and calls, and stores and exports the results. Two more parts belong to the system although they are not in the box: the consumables, which are the plates, tubes and seals, and the assay, the chemistry in the tube.
The instrument's behavior is not a property of any one of these parts. It emerges from how they work together, which is why the most consequential engineering decisions are about the connections between them.
The interfaces are where the system lives
Each pair of subsystems that touch shares an interface: a boundary with parameters that both sides must agree on. The thermal interface between block and plate depends on the shape of the wells, the wall thickness of the plastic and the pressure of the lid. The optical interface depends on the seal or cap the light passes through, its clarity and its own faint fluorescence. The data interface carries run settings one way and results the other, between the instrument and the computer and between the computer and the laboratory's information system. The electrical interface to the mains and to the electromagnetic environment of a laboratory is governed by safety and compatibility standards (Chapter 19). The human interface covers the software screens and the physical loading of plates, and it is where a user can enter the wrong reaction volume and so change the temperature model of Chapter 10 without noticing.
A systems engineer keeps an interface definition for each of these boundaries: what crosses it, in what units, within what limits, and who owns each side. For a PCR instrument the list includes plate geometry and material, seal optics, and lid pressure and temperature. It also includes the volume range the thermal model assumes, the dyes and channels calibrated, the timing of optical reads within a step, the data formats and the version of the analysis software. Most failures that appear when an instrument meets a new assay, plate or laboratory are failures at one of these boundaries.
Choreography
The firmware's central task is timing. Within each cycle, it must bring the calculated sample temperature to each set point and hold it for the programmed time, and at the step the assay specifies it must also trigger the optical subsystem to read every well in every channel the assay uses. In the CDC test the read happened during the 55 °C step, which lasted 30 seconds. The read has to fit inside that step after the liquid has reached temperature. It also has to happen at the same point in every cycle, because probes are still being cut during the step, so a reading taken a few seconds earlier or later would not be comparable with the readings of the cycles around it.
How long a read takes depends on the optical architecture. One manufacturer states that its scanning carriage reads all channels of a 96-well plate in about 12 seconds, or a single channel in about 3, while a camera instrument takes an exposure for each filter pair and changes filters between exposures. Either way, the read time is part of the cycle time and constrains how short the reading step can be. Meanwhile the thermal control loop runs continuously: it reads the block sensor many times a second and adjusts the current in the modules. The instrument is thus one closed loop, the thermal control, with one measurement, the optical read, placed within it by a schedule. The instrument's performance comes from that coordination, not from any single part.
From copies to a result
The most useful way to understand what the instrument does is to follow one quantity, the number of target copies in a well, through every form it takes on its way to a report. The instrument never measures that number directly. It measures something that depends on the number, and then converts the measurement.
Copies of target are made in the well. As copies are made, the polymerase cuts probes bound to them, and each cut probe releases a fluorophore that emits light when excited. The light passes through the seal and an emission filter and strikes a detector. The detector converts photons into electrons, an analog signal that is amplified and converted into digital counts. Those counts are then corrected in several steps. The instrument locates each well in its field of view and subtracts its own background light. It corrects for differences in sensitivity across the plate and separates the contributions of different dyes using their calibrated spectra. Optionally it divides by a reference dye (Chapter 12). What remains is one number per well, per dye, per cycle.
The software subtracts each well's baseline, draws the threshold and finds the fractional cycle at which each curve crosses it, and it then compares that Cq with a cut-off or a standard curve to give a call or a quantity. The call or the quantity is written into a report or sent to a laboratory information system.
Every one of those conversions can add error, and the final number is only as good as the weakest of them. Error in any conversion is why the calibrations of Chapter 12 exist, and why instruments are verified as a whole. The maintenance guide for one instrument family describes a verification run with a reference plate of RNase P, the human gene of Chapter 13, in which the instrument must distinguish wells of 5,000 copies of the gene from wells of 10,000 with 99.7 % confidence.
The two kinds of well hold 5,000 and 10,000 copies, which differ by a factor of two, one doubling, so their Cq values differ by about one cycle. The scatter of Cq values within each group is measured by its standard deviation. For a bell-shaped scatter, 99.7 % of a group's values fall within three standard deviations of its average. As a rough guide, the two groups can be told apart with 99.7 % confidence only if those spreads do not overlap across the one-cycle gap. Three standard deviations on the near side of each group, six in all, must then fit within the one cycle between their averages, so the standard deviation must be no more than about a sixth of a cycle. That single test exercises the thermal uniformity, the optical uniformity, the calibrations and the analysis software together.
Every reported quantity is the end of a chain of conversions: copies to probe cleavage, to light, to electrons, to counts, to corrected values, to a curve, to a Cq, to a call. A fault anywhere along the chain produces a plausible-looking number. The system is trustworthy only if every conversion is characterized and checked, which is what calibration, verification and controls are for.
Software and data are part of the device
Because the result is computed, the software that computes it is part of the measuring instrument. For a diagnostic test the version matters: the CDC test was authorized for one instrument running one version of its software, version 1.4. Changing the algorithm that sets the baseline or finds the crossing changes the Cq values, and therefore potentially the calls. In 2023 the FDA issued guidance on the content of premarket submissions for device software functions. The guidance applies to device software functions, including firmware and software that controls a device, and so it covers the software of a diagnostic instrument (Chapter 19).
The data matter beyond the run that produced them. As Chapter 13 notes, the 2025 MIQE revision asks that raw fluorescence data be exported in vendor-neutral formats, so that a result can be re-analyzed with different software and checked. An instrument that can only report Cq values and keeps its raw data in a closed format makes that re-analysis impossible.
- A real-time instrument is made of thermal, optical, motion, electronic, firmware and software subsystems, with the consumables and the assay also part of the system, and none of these parts knows anything about DNA.
- Its behavior emerges from how the subsystems work together, so the decisions that matter most concern the interfaces between them: plate fit, seal optics, the thermal model's assumptions, read timing, data formats and software versions.
- Firmware runs the thermal control loop and schedules the optical read, so that every well is read at the same point of the specified step in every cycle.
- The number of copies passes through a chain of conversions, to probe cleavage, light, electrons, counts, corrected values, a curve, a Cq and a call, on its way to a report, and each can add error, which is why calibration, verification and controls exist.
From intended use to specification.
+ The questionHow does a need such as "detect a few copies in about an hour" become numbers that the instrument's parts must meet?
The intended use comes first
One sentence in the CDC test's instructions stated what the test was for: the qualitative detection of nucleic acid from SARS-CoV-2 in upper and lower respiratory specimens, in laboratories certified for high-complexity testing. That sentence is the test's intended use. Every instrument and test begins with such a statement, and in a regulated product it is written down before anything is designed. It names what is measured, in what kind of specimen, by whom, in what setting, to inform what decision, and what kind of result is given.
The CDC test's intended use decides much of the system. Qualitative means a cut-off and rules, not standards and a quantity. SARS-CoV-2 is an RNA virus, so nucleic acid from SARS-CoV-2 means handling that keeps the RNA intact and a reverse transcription step (Chapter 20). Respiratory specimens means a sample type with its own inhibitors and variability, and therefore an internal control: a target amplified from every specimen to show that the specimen and the extraction were adequate, and that amplification worked in any well it shares. In the CDC test it is the human gene RNase P, in a well of its own (Chapter 21). High-complexity laboratories means trained staff, a separate extraction step and an instrument that can be fed plates by hand. A test intended for a clinic and run by a nurse in twenty minutes would share the chemistry and almost nothing else.
The V
Systems engineers draw the path from a need to a working product as a V. The picture is known as the V model and is credited to Kevin Forsberg and Harold Mooz in 1991. Down the left side, the need is decomposed: user needs into system requirements, system requirements into requirements for each subsystem, and those into detailed designs. At the bottom the parts are built. Up the right side they are assembled and checked: each part against its own requirements, the assembled subsystems against theirs, the whole system against the system requirements, and finally the finished product against the original user needs.
The two checks on the right side have names that regulators use precisely. The FDA's 1997 guidance on design controls defines verification as confirming that the design's outputs meet its inputs, that the product was built as specified. It defines validation as confirming that the device meets the user's needs and intended uses, that the right product was built. The same guidance asks that design inputs be developed to an engineering level of detail, and it notes that real development loops back between phases. Together these documented steps, from design inputs to verification and validation, are called design controls, and in the United States they are now required through the Quality Management System Regulation, the FDA's rule for the quality systems of device makers. The regulation took effect on 2 February 2026 and incorporates the international quality standard ISO 13485:2016, including its clause on design and development.
Requirements flow down
The left side of the V is where a need such as "detect a few copies in about an hour" becomes numbers. Each system requirement splits into requirements on several subsystems, and most subsystems answer to several system requirements. The example below is illustrative and is built from figures in earlier chapters.
Sensitivity. A requirement to detect about five copies per reaction, the figure of Chapter 1, sits just above the sampling floor of about three copies, below which no limit can go (Chapter 6). The assay must therefore put enough of the specimen into each reaction: in the CDC test, 5 microliters of extract in a 20-microliter reaction. The extraction must recover the RNA efficiently and remove inhibitors. The optics must distinguish the light of 10¹⁰ to 10¹¹ copies of product from the background reliably enough that the threshold can sit low. The run must also have enough cycles for a handful of copies to cross that threshold at realistic efficiency. In the CDC test the cut-off was at cycle 40 of 45.
Time to result. The run time is the number of cycles times the cycle time, plus the initial steps. The cycle time is the hold times plus the time the liquid takes to move between temperatures twice, plus any time the read adds. So the time requirement becomes a requirement on the sample ramp rate, on the read time and on the sample-temperature model of Chapter 10, with which the controller brings the liquid itself to each set point so that a short hold is a real one.
Throughput. A requirement for a number of specimens per run becomes a plate format and a layout. If each specimen takes three wells, as in the CDC test from March 2020, and three controls take three wells each, a 96-well plate holds 29 specimens. A multiplex test that detects all targets in one well with different dyes needs one well for each specimen and each control, so it would hold 93, about three times as many, at the cost of more optical channels, more calibration and the separation of overlapping dye signals (Chapter 12).
Specificity and robustness. The assay's tolerance to temperature error becomes a requirement on accuracy and uniformity at the annealing and denaturing steps, and the instrument's guaranteed accuracy and uniformity become a requirement on the assay's robustness.
The CDC test's cycle holds 3 s at 95 °C and 30 s at 55 °C, so the liquid swings 40 degrees each way. The instrument's maker states in its brochure that the sample ramps at 1.6 °C per second in standard mode and 3.5 °C per second in fast mode. In standard mode each ramp takes 40 ÷ 1.6 = 25 s, so a cycle takes about 3 + 30 + 25 + 25 = 83 s, and 45 cycles take about 62 minutes. In fast mode each ramp takes 40 ÷ 3.5, or about 11 s, so a cycle takes about 56 s and 45 cycles about 42 minutes. The TaqPath mix adds 19 minutes of initial holds, so the run takes about 81 minutes in standard mode or about 61 in fast mode, not counting the read time. In standard mode the two ramps take 50 of the 83 seconds, more of the cycle than the holds, while in fast mode the holds take the larger share. The test's instructions required the standard mode, so that is how the test had to run.
Error budgets
Where a requirement depends on several contributions, engineers allocate it among them in an error budget. Temperature at the annealing step is a clear case. The actual temperature of a given well differs from the set point by the instrument's accuracy, plus that well's departure from the average, plus any residual error from the sample-temperature model, plus the uncertainty of the sensor's calibration.
One manufacturer specifies ±0.2 °C accuracy and ±0.4 °C uniformity within 10 seconds of arrival, both stated at 90 °C. Those two figures alone put a well on that instrument up to about 0.6 °C from its set point at 90 °C before the model and calibration errors are counted, and the specification does not say what holds at the annealing step. The maker of the CDC's instrument gives ±0.5 °C accuracy and ±1 °C uniformity in its brochure, figures that allow about 1.5 °C. The brochure does not say at what temperature or how soon after arrival they were measured, so the two makers' figures cannot be compared directly. Engineers often combine independent errors by adding their squares and taking the square root, which gives a smaller, more likely figure, but a diagnostic test must work in the worst well.
The temperature budget closes on the assay side. A test developer establishes how far the annealing temperature can move before the test loses sensitivity or specificity, typically by running the test deliberately at temperatures above and below the set point. That tolerance must exceed the instrument's worst-case error with margin. The optical budget closes the same way. Background, noise and calibration error set the lowest signal the instrument can distinguish, and that signal sets where the threshold can sit. The position of the threshold in turn sets how precisely Cq is determined, and that precision is what the verification plate of Chapter 15 checks.
Trade-offs
Requirements pull against each other, and the systems engineer's job is to make the trade explicit rather than let it happen by accident. Speed pulls against uniformity, because a lighter block changes temperature faster and spreads heat less evenly. More optical channels allow more targets per well, but they cost more, demand more calibration and leak more light between channels. More cycles reach lower copy numbers but lengthen the run. A sealed cartridge prevents contamination and simplifies the user's work, but it fixes the assay and the volume. Every one of these trade-offs is a decision about which requirement matters more for the intended use, and every one shows up later as a specification someone must verify.
Each figure in an instrument specification should trace up to a system requirement and from there to the intended use, and down to a verification method that will show it is met. That holds for a ramp rate, a uniformity, a read time and a channel count alike. A number that traces to nothing is either unnecessary or a sign of a requirement nobody wrote down.
- The intended use states what is measured, in what specimen, by whom, where and for what result, and it decides much of the system before any design begins.
- The V model decomposes needs into system and subsystem requirements and then checks upward, with verification showing that the product meets its specifications and validation showing that it meets the user's needs.
- Sensitivity, time, throughput and robustness each flow down to several subsystems, and in the CDC test's standard mode the ramps take more of each cycle than the holds do.
- Error budgets allocate a requirement among its contributors and close against the assay's tolerance, and trade-offs between speed, uniformity, channels and cycles are made explicitly.
Where the instrument meets the chemistry.
+ The questionWhy can't an instrument be specified without its assay, or an assay validated without its instrument?
The tube is an interface
The plastic tube or plate looks like packaging. It is an engineered part of both the instrument and the test, and it sits on two interfaces at once (Chapter 15).
On the thermal side, the wall of the tube is the resistance that makes the liquid lag behind the block (Chapter 10). A thinner wall, a smaller volume or a lower-profile well shortens the lag, which is why fast instruments use their own plates. The maker of the instrument named in the CDC test states in its brochure that the instrument takes fast 96-well plates or 8-tube strips with 100-microliter wells, and reactions of 10 to 30 microliters. The tube's fit in the block and the lid's pressure on it decide how well heat crosses from metal to plastic. The volume in the tube is an input to the instrument's sample-temperature model, so a change of plate can change the temperature the liquid actually reaches without any change to the instrument's settings.
On the optical side, light enters and leaves through the cap or seal. The seal's clarity, its own faint fluorescence and any condensation on it all affect the signal, and preventing that condensation is one reason for the heated lid. The reaction volume sets the height of the liquid in the well and therefore how much of the liquid the optics see.
The chemistry is matched to the machine
The reagents are matched to the instrument as closely as the plastic is. A master mix's hot-start enzyme needs an activation step of a certain length at a certain temperature, and the instrument's profile must provide it. The CDC test's four authorized mixes used three different initial steps. Reverse transcription ran for 15 minutes at 50 °C, 10 minutes at 50 °C or 15 minutes at 45 °C, followed by 2 or 3 minutes at 95 °C.
The CDC test's instructions set the initial steps for each authorized mix before the 45 cycles. The TaqPath mix holds 2 minutes at 25 °C, 15 minutes at 50 °C and 2 minutes at 95 °C, so its initial steps take 2 + 15 + 2 = 19 minutes. Two of the other mixes share initial steps of 10 minutes at 50 °C and 3 minutes at 95 °C, 13 minutes in all. The fourth holds 15 minutes at 45 °C and 2 minutes at 95 °C, 17 minutes in all. The 45 cycles are the same for every mix and take about 62 minutes in standard mode, using the ramp rate the instrument's maker states (Chapter 16). A run therefore takes about 81 minutes with TaqPath, about 75 with either of the pair and about 79 with the fourth, before any read time. The choice of mix alone changes the run by up to 6 minutes, and the instrument's profile must provide each mix's steps.
A mix formulated for slow cycling may not finish its extension in the short holds of a fast instrument. The fluorophores on the probes must also be ones the instrument has channels for and has been calibrated with. The brochure for the CDC's instrument lists the dyes the instrument was calibrated for, beginning with FAM, the dye that all three of the test's probes carried.
The software is part of the method
Then there are the settings. They look like configuration and are in fact part of the test. The CDC's instructions specified that the instrument run in its standard mode, not its faster mode, and that the passive reference dye be set to none. In the software each target is entered as a detector, and the instructions specified that each detector be set to FAM with the quencher set to none (Chapter 12). They required the baseline to be left at the default and the threshold to be set by hand in the exponential region. Under the instructions, a target counted as detected below a Ct of 40.00.
Each setting changes what the instrument does or how its output is read. Running the same chemistry in fast mode would change the time the liquid spends at each temperature. Changing the threshold would change every Ct, and changing the cut-off would change the calls.
In a validated test, the instrument's run mode, reference-dye setting, analysis settings and software version are part of the method whose performance was established. Any change to them, however reasonable it looks, means that the established performance no longer applies until it has been shown again.
The test system
Regulators draw the consequence explicitly. What they evaluate is a test system: the assay, the instrument and the software together. The FDA finalized its guidance on assay migration in April 2013. The guidance covers moving an assay to a new instrument system whose performance the agency has not evaluated, and it defines the system as the assay, the instrument and the software. It is written for approved class III assays, licensed donor-screening tests and cleared assays whose migration raises particular concerns. Approval, licensing and clearance are separate FDA routes to market, and Chapter 25 explains the device classes and clearance.
For a quantitative test the guidance recommends, among other studies, testing precision at several sites, comparing results on at least 180 samples across the measuring range and showing that the limits of blank and detection are equivalent (Chapter 21). For a qualitative test it recommends panels of low-level samples and comparison panels of at least 100 positive and 100 negative samples. For every nucleic acid test it recommends studies of carry-over between samples.
The FDA finalized a later policy in August 2022, and that policy starts from the premise that an assay is cleared for use on specific instruments. It describes an instrument family: instruments from the same manufacturer with the same general architecture, design, tolerances and capabilities. It also describes the conditions under which a cleared assay may be extended to another member of the family without a new submission. The policy notes that a new submission is likely to be needed if the software must be modified to run the assay on the added instrument.
The CDC test is an example of a test system in practice. Its instructions named one instrument, with one software version. They named the master mixes that could be used: initially one, with three more added on 30 March 2020. They named the extraction methods: a list of kits and automated systems, extended in June and July 2020 as new platforms were shown to work. They also named a heat-treatment alternative, with a warning that it had been tested on few clinical specimens and might lose sensitivity. Each addition came with data and a revised authorization. A laboratory that ran the test on another instrument, or with another extraction kit, was no longer running the authorized test. It was running a modified test whose performance it had to establish itself.
What this means for developers
For a company developing an assay, the choice of instrument platform is therefore a long-term commitment: the assay's performance will be established on that platform, and moving the assay later is a project with its own studies. For a company developing an instrument, the assays that will run on it are part of its design inputs: the plates, volumes, dyes, profiles and analysis settings they need, and the stability of those parameters over the instrument's life. Software updates, a new lamp or LED supplier, a changed block design or a different plate supplier all affect the tests that run on the instrument, so each change must be controlled with those tests in mind.
Some instruments are closed: they run only the manufacturer's own assays, in its own consumables, with fixed settings, so that the test system is defined once. Others are open: they run any assay a laboratory develops. That openness offers flexibility, and it moves the burden of establishing the test system's performance onto the laboratory.
Once an assay's performance is established on an instrument, write down the complete test system: instrument model and software version, plate and seal part numbers, reaction volume, master mix, profile, run mode, dye and analysis settings, extraction method, controls and interpretation rules. Treat any change to any item as a change to the test. Assess it, and show with data that performance still holds before the changed test is used.
- The plate or tube sits on both the thermal and the optical interface, so changing it can change the temperature the liquid reaches and the light the optics see.
- Reagents are matched to the instrument's profile, speed and calibrated dyes, and run settings such as mode, reference dye, threshold and cut-off are part of the method.
- Regulators evaluate the assay, the instrument and the software together as a test system, assays are cleared for specific instruments, and moving an assay to a new instrument calls for migration studies or an instrument-family justification.
- The CDC test's instructions named one instrument, one software version, specific mixes and extraction methods, and each addition needed data and a revised authorization.
The lab in a cartridge.
+ The questionWhat does it take to put extraction, PCR and reading into one closed cartridge?
Why put a laboratory in a cartridge
A test like the CDC's runs in a molecular laboratory. A trained technician extracts nucleic acid from each specimen with a kit or a separate automated system, then pipettes the extract and the reagents into a plate, seals it, loads it into the instrument and, after the run, interprets the curves. In February 2020 CDC estimated that the process took about four hours from the start of sample processing to a result, and the test was authorized only for laboratories certified for high-complexity testing. Every manual step needs skill, takes time and opens a route for error or contamination. Every opened tube of specimen or extract is a route for the cross-contamination that causes false positives, although the real-time plate itself stays sealed after cycling (Chapter 11).
A sample-to-answer system puts all of those steps into one sealed disposable cartridge and one instrument. The user adds the specimen, closes the cartridge, puts it in the instrument and receives a result. For one SARS-CoV-2 test cartridge, authorized in March 2020, its maker states about 45 minutes to result, and an FDA decision summary gives 15 to 20 minutes for another system. The maker of a third system, which runs a panel of 22 respiratory targets at once, states about 45 minutes. The engineering that makes this possible is integration: every function of the laboratory is built into a molded part that costs little enough to throw away.
A sample-to-answer test uses its own extraction, its own reaction volume and its own targets. Its performance must be established on its own terms against a reference, not assumed from a laboratory test with the same name. Its specimen volume, extraction, reaction volume and cycle number all differ from the laboratory test's, and any of them can change its limit of detection.
One cartridge, one valve
The GeneXpert system shows one way to integrate the steps. It was described in its maker's technical documents and in a 2005 paper whose authors included four staff of the maker, Cepheid, alongside scientists at the University of Pittsburgh. Its cartridge is a plastic body with chambers for the sample, the reagents and the waste, arranged around a central rotary valve. The instrument turns the valve to connect any chamber to a central channel, and it drives a plunger that draws liquid in and pushes it out, so that one moving part routes every liquid where it is needed, in sequence. An ultrasonic horn in the instrument presses against the cartridge to break open cells and viruses.
The reagents are freeze-dried beads that dissolve when liquid reaches them: an enzyme bead, a bead of primers and probes for the target, and a bead containing a sample-processing control. The final reaction mixture is pushed into a thin reaction tube, which sits both in a thermal module with ceramic heater plates and a fan for cooling and in optical blocks that read six or more colors. The maker states that no part of the instrument touches the sample, so that nothing carries over from one cartridge to the next.
The authors of the 2005 paper reported that the system extracted RNA in just over six minutes, with about 70 % of the yield of a conventional laboratory kit. They reported that it completed extraction, reverse transcription and real-time PCR in about 35 minutes, with 95 % of replicate Ct values expected within about 1.5 cycles either side of the average. The system's SARS-CoV-2 test was authorized in March 2020. It detected two targets and carried two built-in checks: a sample-processing control that showed extraction and amplification worked, and a probe-check control that verified the dried reagents had dissolved, the reaction tube had filled and the probes were intact before cycling began. The test could also stop early and report a strongly positive sample before the full run finished.
Two stages in a pouch
The FilmArray system takes a different route and runs many tests at once. It was described in 2011 by scientists at Idaho Technology, the company that developed it, with colleagues at the University of Utah, and the company paid for part of the work. Its pouch has a rigid plastic fitting with reservoirs for reagents and a flexible film with blisters in which the work happens. The instrument moves liquid between blisters by pressing on them with pneumatic actuators. Ceramic beads beaten against the sample break open the organisms. Magnetic beads coated with silica then capture the nucleic acid, are washed and release it.
A first stage of PCR amplifies all the targets together in one blister for 26 cycles of 4 seconds at 94 °C and 19 seconds at 60 °C. The product is then diluted into an array of 102 wells of about one microliter each, and every well is pre-loaded with primers for one target. A second stage of 30 cycles runs inside each well with primers that bind within the first-stage product, a design called nested PCR that adds specificity. A dye and a melt curve in each well identify the products (Chapter 12). Two Peltier devices, the pneumatic actuators, a blue LED and a camera are the instrument's essential hardware. In the 2011 description the whole process took about an hour; the maker states about 45 minutes for the panel of 22 respiratory targets it sells as of October 2026.
The 2011 description gives the holds of both stages. The first stage runs 26 cycles of 4 seconds at 94 °C and 19 seconds at 60 °C, which is 23 seconds of holds per cycle: 26 × 23 = 598 seconds, or about 10 minutes. The second stage runs 30 cycles of 4 seconds at 94 °C and 19 seconds at 63 °C, also 23 seconds per cycle: 30 × 23 = 690 seconds, or about 11.5 minutes. The final melt heats the wells from 68 to 95 °C at 0.5 °C per second, which takes 27 ÷ 0.5 = 54 seconds. Holds and melt together take 598 + 690 + 54 = 1,342 seconds, about 22 minutes. The other 38 minutes or so of the hour go to everything else, including breaking open the organisms, capturing and washing the nucleic acid, moving and diluting the liquid, and the ramps between temperatures.
What integration costs
Putting a laboratory into a cartridge moves the difficulty rather than removing it. Liquids must be moved reliably in small volumes without bubbles, without leaks and without leaving residue that carries over between steps. Reagents must survive storage at room temperature, which is why they are freeze-dried. They must also dissolve completely and uniformly at the right moment. The thermal design must suit a small volume in a thin chamber, and the optics must read through molded plastic. Controls must be built into every cartridge, because no technician watches each step. And the cartridge must be manufactured in large numbers at low cost with every chamber, valve and seal within tolerance, because one bad cartridge is one wrong result.
Point-of-care testing is done near the patient rather than in a laboratory. In a 2012 review of point-of-care microfluidic devices, the authors observed how few of the many technologies demonstrated in research had reached the market, and they argued for designing integrated, real-world systems rather than optimizing single components. Integration is where these products succeed or fail.
Not every fast molecular test is PCR. The maker of one point-of-care COVID-19 test states that it gives results in 13 minutes or less, and the test amplifies its target with an isothermal method, at a single temperature, with no thermal cycling at all. Chapter 23 explains what such methods give up in exchange for removing the cycle.
In a sample-to-answer product, the cartridge is most of the system. Its fluid paths, valve timings, reagent formats, thermal chamber, optical window and built-in controls are designed together with the instrument that drives them, and verified together, with manufacturing tolerances included. Treating the cartridge as a consumable to be specified after the instrument is finished leaves the hard problems until last.
- Sample-to-answer systems put extraction, amplification, detection and controls into one sealed cartridge, cutting hands-on steps, contamination risk and time to result.
- One design routes all liquids with a single rotary valve and a plunger, breaks open cells and viruses with ultrasound, and uses freeze-dried reagent beads with built-in process and probe controls.
- Another design runs a first-stage multiplex PCR and a nested second stage in an array of about 100 one-microliter wells in a pneumatically driven pouch.
- Integration moves the difficulty into fluidics, reagent stability, small-volume thermal design, built-in controls and manufacturing tolerance.
Verifying the system.
+ The questionHow do you show that every part, and the whole, does what it was specified to do?
Two questions on the right side of the V
A thermal block that runs cold in one corner can make a specimen with little virus cross the threshold too late, or not at all, and the result is then a false negative. Faults of that kind are what verification looks for.
The right side of the V in Chapter 16 asks two different questions. Verification asks whether the product was built as specified: does each part, and the assembled whole, meet its written requirements? Validation asks whether the specified product is the right one: does it meet the user's needs in the intended use? For a PCR instrument and its test, verification runs from the smallest part up to the integrated system, and validation shows that the test gives correct results on real specimens. The two meet in the analytical studies of the assay on the finished system, which belong to both.
Subsystems first
Each subsystem is verified against its own requirements, by methods that measure the property directly.
The thermal subsystem is measured in the wells with traceable probes (Chapter 10). The measurements cover accuracy and uniformity at each set point, overshoot and undershoot, stability during holds, and ramp times under the conditions the specification states, with the plates and volumes the assays will use. The optical subsystem is verified for its ability to locate each well, its background, its uniformity across the plate, its response to known amounts of each calibrated dye, and the separation of dyes in neighboring channels. Motion is verified for positioning and repeatability, the electronics for noise and for the drive currents the modules need, and the firmware for the timing of events within each cycle, so that the read falls where it should every time.
Software is verified as carefully as hardware, because it computes the result. In June 2023 the FDA finalized its guidance on the content of premarket submissions for device software functions. The guidance applies to device software, including firmware and software that controls a device, and it asks for documentation at one of two levels: a basic level, or an enhanced level where a software failure could present a probable risk of death or serious injury. It also allows a manufacturer to declare conformity with IEC 62304, the recognized standard for the software life cycle, as part of that documentation.
The current edition of IEC 62304 dates from 2006, with an amendment in 2015. A second edition is in development and, as of October 2026, was not expected before 2028. In vitro diagnostic devices are the tests and instruments that examine specimens taken from the body, and in the European Union they fall under the IVD Regulation (Chapter 26). Its general safety and performance requirements include requirements for software, and a manufacturer can use IEC 62304 to show that its software meets them.
Then the system
Assembled subsystems can pass every test and still fail together, which is why the integrated instrument is verified as a whole. The reference-plate test of Chapter 15 shows the kind of whole-system test needed. It is an installation and periodic check, and its one plate of 5,000 and 10,000 copies of RNase P exercises the thermal, optical, calibration and software parts together.
Beyond the reference plate, the assay's analytical studies are run on the finished system. They cover the limit of detection and the precision across runs, days, operators, reagent lots and instruments. They also cover the ability to tell targets from near relatives, and carry-over: whether a strongly positive sample can contaminate the negative sample run after it. The FDA's guidance on moving assays between instruments asks for carry-over studies in every migration of a nucleic acid test for exactly that reason (Chapter 17).
Subsystems verified separately can still fail together: a plate that fits the block but fogs under the lid, an optical read that drifts into the ramp when the firmware timing changes, a software update that moves the baseline. The integrated system, with its consumables and its assays, must be verified as a whole, and re-verified when any part of it changes.
Risk: how a wrong number becomes harm
A diagnostic instrument rarely hurts anyone directly. It can harm through its result: a false negative that sends an infected person home untreated, a false positive that leads to an unnecessary treatment or isolation, a wrong quantity, a result that arrives too late to matter. Risk management for a diagnostic device therefore traces each possible fault forward to a wrong result and on to its consequence. ISO 14971:2019 is the international standard for medical device risk management, and it applies to diagnostic devices; ISO confirmed it in March 2025. Its companion guidance, ISO/TR 24971:2020, includes an annex on applying the standard to diagnostic devices, and in July 2026 ISO marked that guidance for revision.
A worked chain, in ISO 14971's terms, shows the method. For a diagnostic device the hazard is usually taken to be an incorrect result. A sequence of events starts with a fault: the block runs cold in one corner at the denaturing step, some target strands in those wells do not separate, and amplification is delayed or fails, so a specimen with little virus crosses the threshold too late or not at all, giving a false-negative result. A hazardous situation: that result is reported and acted on. A harm: a missed diagnosis.
Controls can act at each step. The design adds edge heaters and guard zones. Verification measures the corners at installation and at intervals. Detection catches gross failures. The positive control fails if it sits in an affected well. Every specimen is also tested for its internal control, a human gene (Chapter 16). If the internal control's own well is among the cold ones, a specimen whose control is not detected before the cut-off is reported invalid rather than negative, but a modest delay that leaves the internal control inside the cut-off passes unnoticed.
The last control is information for the user: the labeling states the limits of a negative result. For multi-target tests that include SARS-CoV-2, the FDA sets special controls: requirements that every device of that type must meet (Chapter 25). Under them, the labeling must state that a negative result does not rule out infection and that sequence variants can cause incorrect results. The special controls also require continuous monitoring for new variants, and results suggesting that a variant affects the stated performance must be sent to the agency immediately, with the evaluations submitted within 48 hours if the FDA asks for them.
In the CDC test as authorized from March 2020, each specimen takes three wells, one each for N1, N2 and the internal control, RNase P. The run's three controls take three wells each, which is 3 × 3 = 9 wells. A 96-well plate then has 96 − 9 = 87 wells left for specimens, and at three wells each it holds 87 ÷ 3 = 29 specimens, as in Chapter 16. The specimens' internal controls fill 29 of the plate's 96 wells, about 30 %. Almost a third of a full plate therefore goes to the internal control.
In the CDC test, a specimen with no viral target is reported negative only if the human control gene, RNase P, is detected (Chapter 13). That rule converts several silent failures into an invalid result, which prompts a repeat, instead of a false negative, which does not. The failures include a swab with too few cells, a failed extraction and an inhibitor that stopped the reaction. A well that never reached temperature is caught only when the control shares that well, as in a multiplex test, and in the CDC test RNase P ran in a well of its own (Chapter 21). One extra target per specimen turns a hazard that would otherwise pass unnoticed into one the system detects.
Safety, electromagnetic compatibility and the user
A laboratory instrument must also be safe to use and compatible with the other electrical equipment around it. The IEC 61010 family of standards covers the safety of electrical equipment for measurement, control and laboratory use, against shock, fire, heat and mechanical hazards. Part 1 gives the general requirements. Its edition of 2010 was amended in 2016 and consolidated in 2017, and a second amendment had completed its final vote in September 2026 but had not been published as of October 2026. Particular parts add requirements: Part 2-010 for equipment that heats materials, Part 2-081 for automatic laboratory analyzers, and Part 2-101, from 2018, for in vitro diagnostic equipment. Which parts of IEC 61010 apply to a given instrument is a judgment made from its functions. For example, one manufacturer states that its air-heated cycler conforms to Parts 1 and 2-101.
Electromagnetic compatibility is the instrument's ability to work without disturbing other equipment or being disturbed by it. IEC 61326-1 covers it, and a particular part for diagnostic equipment, IEC 61326-2-6, was published in a new edition in 2025. IEC 62366-1 covers usability engineering, which addresses the errors users can make, such as entering the wrong volume or loading a plate the wrong way round, and the labeling standard ISO 18113-3 covers the information supplied with diagnostic instruments for professional use.
Production and the field
Verification does not end when the design is approved. Every unit built is tested before it ships, with its thermal and optical performance checked against limits, and each unit is calibrated at installation and checked at intervals in use. One accredited calibration service recommends calibration once a year or every 250 runs. The Peltier modules, the fan, the lamp and other parts with finite lives are inspected or replaced on a schedule. Complaints and service records are collected and analyzed for trends.
All of this happens inside a quality management system. For medical devices the international standard for such a system is ISO 13485:2016, which was confirmed without change after review in October 2025. In the United States, the Quality Management System Regulation of Chapter 16 has been in force since February 2026 and incorporates ISO 13485 (Chapter 25). On the day the regulation took effect, the FDA stopped using its Quality System Inspection Technique and began inspecting under a new compliance program.
The thermal mapping, optical verification, software documentation, risk file, safety and compatibility test reports, analytical studies and quality records described here are one body of evidence. When it is organized around the system's requirements from the start, the same core serves design reviews, customers' audits, the investigation of field problems and regulatory submissions in several jurisdictions (Chapters 25 and 26).
- Verification shows that the product was built as specified, from subsystems up to the integrated system, while validation shows that it meets users' needs, and analytical studies on the finished system belong to both.
- Thermal, optical, motion, electronic, firmware and software subsystems are each verified directly, and software documentation follows the FDA's 2023 guidance, which allows a declaration of conformity with IEC 62304.
- A diagnostic instrument harms through wrong or late results, so risk management traces faults to wrong results and adds controls at each link, and the internal control turns several silent failures into invalid results.
- The safety standards of IEC 61010, the compatibility standards of IEC 61326 and standards for usability and labeling apply, and verification continues in production and the field under a quality system.
+ Part VI · The test
From a specimen to a trusted result.
The CDC's test had a limit of detection of about 3 RNA copies per microliter of a sample spiked with a known amount of the target's RNA, with one automated extraction system, and of 1 copy per microliter with a manual kit. The step before the reaction set part of what the test could detect. An instrument and an assay that work in the development laboratory are not yet a test. This part follows the steps that make them one: reading RNA with a method that copies DNA, getting nucleic acid out of a specimen intact and free of inhibitors, the controls and limits that make a result believable, and the CDC's test as it was designed, made and shipped in the first weeks of 2020.
Reading RNA with a DNA method.
+ The questionMost of the viruses that matter carry RNA. How does a method that copies DNA read them?
A step before the first cycle
Taq polymerase copies DNA. It needs a DNA template, and it does not copy RNA usefully. Yet many of the targets that matter most in diagnostics are RNA: the genomes of SARS-CoV-2, influenza, respiratory syncytial virus, HIV and hepatitis C virus are all made of RNA. The activity of a gene, too, is measured through its messenger RNA, the RNA copy that a cell makes of a gene in order to use it. Reading these targets with PCR needs one extra step at the start: copying the RNA into DNA.
The enzyme that does this was discovered in 1970 in retroviruses, viruses whose RNA genomes must be copied into DNA to infect a cell, and two teams found it independently: Howard Temin with Satoshi Mizutani, and David Baltimore. The enzyme is called reverse transcriptase, because it runs the usual flow of genetic information backward, from RNA to DNA. Given an RNA template and a primer, it builds a strand of complementary DNA, or cDNA, which PCR can then amplify. The enzymes used in laboratories come from two other retroviruses, avian myeloblastosis virus and Moloney murine leukemia virus, or are engineered versions of those enzymes. They work at moderate temperatures, around 40 to 50 °C for the original enzymes, and a higher temperature helps them read RNA that folds back on itself.
One tube or two
The reverse transcription and the PCR can be run in separate tubes or in one. In a two-step method, the RNA is first copied into cDNA in its own reaction, with its own buffer, and portions of that cDNA are then used in separate PCRs. Each step can be optimized, and the cDNA can be stored and used for several tests. In a one-step method, both enzymes and both sets of conditions share one tube: the reaction is held at the reverse-transcription temperature for some minutes, then heated, and the PCR cycles begin. Fewer steps mean less handling and less risk of contamination, which is why one-step methods suit diagnostic testing.
The reverse transcriptase needs a primer too. In a one-step diagnostic test, the PCR's own reverse primer usually serves: it binds the viral RNA and is extended into cDNA, which the forward primer then copies in the first PCR cycle. The CDC test was one-step. With the TaqPath master mix, its profile began with 2 minutes at 25 °C, a step for the contamination control described at the end of this chapter. Then came 15 minutes at 50 °C for reverse transcription and 2 minutes at 95 °C to activate the polymerase, before the 45 cycles began.
With the TaqPath master mix, the CDC test's profile held 2 minutes at 25 °C, 15 minutes at 50 °C and 2 minutes at 95 °C, so 2 + 15 + 2 = 19 minutes passed before the first cycle. Each of the 45 cycles then held 3 seconds at 95 °C and 30 seconds at 55 °C, or 33 seconds. Over the run that makes 45 × 33 = 1,485 seconds, about 25 minutes. Counting only the time at set temperatures, and leaving out the time the block spends changing temperature, the holds came to about 19 + 25 = 44 minutes, while the whole run, ramps included, took about 81 minutes in standard mode before any read time (Chapter 16). Reverse transcription took 15 of those 44 minutes, so reading RNA cost about a third of the time this run spent at its set temperatures, and a little under a fifth of the whole run.
The test's two viral targets lay in the gene for the nucleocapsid protein, the protein that packages the viral RNA, and this gene is called the N gene. In their paper describing the test, the CDC's scientists gave one reason for the choice. During infection, the virus makes many extra RNA copies of its N gene region, and these copies may make the target more abundant than other regions and the test more sensitive.
When RNA is the target, any DNA in the sample with the same sequence will also be amplified, with or without reverse transcription. A reaction run without the reverse transcriptase, called a no-RT control, shows whether the signal came from RNA or from contaminating DNA. The 2025 MIQE revision asks for such controls in studies that measure gene activity with reverse transcription and real-time PCR, where the gene's DNA is present in every cell.
RNA is fragile, and enzymes that destroy it are everywhere
RNA is easier to destroy than DNA. The enzymes that destroy it are called ribonucleases, or RNases, and they are everywhere: in tears, saliva, mucus and sweat and on skin and hair, and bacteria and fungi make them too. One supplier of laboratory reagents describes RNases in a technical note as small, compact proteins held together by several internal cross-links. The same supplier states that they are robust enough to refold after being heated and to keep substantial activity even after autoclaving, which sterilizes laboratory equipment with steam under pressure.
A specimen can lose its target RNA before the test begins: in the swab itself, in a tube touched with a bare hand, or through a pipette tip that was not certified free of RNase. These are the reasons for gloves, certified plastics, inhibitors of RNase in the reaction mix and cold storage of specimens.
Getting the nucleic acid out
Between the swab and the reaction comes extraction, the step that recovers the nucleic acid from the specimen. Its jobs are to break open the virus and the cells, to separate the nucleic acid from everything else in the specimen, including RNases and substances that would slow the polymerase, and often to concentrate it. The most common method breaks open the particles with a strong salt solution of guanidinium, which also unfolds, or denatures, the proteins. The nucleic acid is then bound to silica, on a small column or on magnetic beads. The silica is washed with buffers containing ethanol, and the purified nucleic acid is released in a small volume of water or buffer.
Extraction is part of the test's performance, not a preliminary to it. The CDC test's instructions listed the extraction kits and automated systems that the test was validated with. The test's limit of detection differed between them: about 3 RNA copies per microliter of the spiked sample with one automated system, and 1 copy per microliter with a manual kit. A spiked sample is one into which a known amount of the target's RNA has been mixed. The instructions also allowed a simple heat treatment in place of extraction, with a warning that the heat treatment had been tested on few clinical specimens and that some loss of sensitivity could not be ruled out.
Extraction can also add a problem of its own. A 2020 study measured how traces of the extraction kits' buffers, carried into the reaction, delay amplification. Ethanol wash buffer making up about 10 % of the reaction delayed the Cq by at least half a cycle, and so did as little as 0.3 to 1 % of the guanidinium buffer that breaks the particles open.
Inhibitors
Many substances in clinical and environmental samples slow or stop the reaction. Such substances are inhibitors, which lower the efficiency of amplification (Chapter 13). The authors of one review classify them by where they act: on the nucleic acid, on the polymerase, on primer binding, or on the fluorescence itself.
| Inhibitor | Typical source | How it interferes |
|---|---|---|
| Hemoglobin and heme | red blood cells | lowers polymerase activity; can quench fluorescence |
| Lactoferrin | white blood cells | releases iron; interacts with nucleic acids |
| Immunoglobulin G | plasma | binds single-stranded DNA, hindering primer binding |
| Heparin, EDTA | blood collection tubes | heparin inhibits the reaction; EDTA removes the magnesium the polymerase needs |
| Humic and fulvic acids | soil and sediment | act on the polymerase; quench DNA-binding dyes |
| Ethanol, guanidinium | carry-over from extraction kits | delay amplification |
An inhibitor lowers the efficiency, so the curve rises later, and a strong enough inhibitor keeps it from rising at all. In a test without an internal control, a reaction stopped by an inhibitor looks exactly like a negative result. Additives can help. In a study of inhibitors from blood, the most effective additive at relieving them was bovine serum albumin, a protein purified from cattle blood.
A specimen that inhibits PCR gives no signal for the target, just like a specimen without the target. Only a control amplified from the same specimen distinguishes the two. When the control shares the target's well, it also shows that this very reaction could amplify. Telling the two apart is the job of the internal control described in Chapter 21. It is also why a specimen with no signal for the target and none for the internal control is called invalid rather than negative.
Keeping old product out of new reactions
Every positive PCR makes billions of copies of its target. A laboratory that runs thousands of reactions accumulates those copies on benches, pipettes and gloves, and a trace of old product in a new reaction is a perfect template that gives a false positive. Laboratories separate the rooms where reactions are prepared from the rooms where amplified product is handled, and real-time reactions help as well, because their tubes stay closed and are never opened after amplification.
There is also a chemical defense, described in 1990 by scientists at Life Technologies, a supplier of products for life-science research. Every PCR is run with dUTP, a building block carrying uracil, in place of dTTP, the usual building block carrying thymine, so that every product contains uracil instead of thymine. Before each new PCR, an enzyme called uracil-DNA glycosylase, or UNG, is allowed to act. It removes uracil from any DNA, and removing the uracil damages carried-over product so that it can no longer be copied. The specimen's own DNA contains thymine, so the enzyme leaves it untouched. The enzyme is then inactivated by heat, before the new uracil-containing copies are made. In the CDC test's TaqPath profile, the first 2 minutes, at 25 °C, are that step.
For a test developer, the extraction method, the specimen types, the RNase controls and the contamination defenses are part of the test system as much as the primers. Their performance must be established with the rest of the test. Any change to them is a change to the test: a new extraction kit, a new transport medium for the swabs, or a heat step instead of extraction.
- Reverse transcriptase, discovered in 1970, first copies an RNA target into complementary DNA, either in a reaction of its own or, in the one-step methods that suit diagnostic testing, in the same tube as the PCR, often with the PCR's reverse primer as the primer.
- RNases are robust and everywhere, so RNA needs careful handling, and extraction releases and purifies the nucleic acid and sets part of the test's sensitivity.
- Inhibitors from the specimen or the extraction slow or stop the reaction, and without an internal control an inhibited reaction cannot be told apart from a negative.
- Carried-over product is kept out by separate work areas and by dUTP with UNG, which damages old product before each run so that it cannot be copied.
Controls, limits and false results.
+ The questionWhat turns a reaction that works into a test whose result can be trusted?
Every result carries its own evidence
A single PCR that gives a signal shows only that something amplified. A test result must say more: that the right target amplified, that nothing else did, that a negative is a real absence and not a failure, and that the test would have detected the target at the concentrations that matter. Each of those claims rests on a specific piece of evidence, some of it built into every run as controls and some established once, before the test is used, as the test's performance characteristics.
Controls: one question each
Controls are reactions, or targets within reactions, whose expected result is known, and each answers one question about the run. A positive control contains the target at a known, modest level (Chapter 8). If it fails, the reagents or the instrument failed, and no negative in that run can be believed. A no-template control contains everything but the sample (Chapter 5). If it amplifies, something in the run is contaminated, and no positive can be believed.
An extraction control is a specimen-like sample carried through extraction alongside the patients' specimens. In the CDC test it was a human specimen control, which had to be negative for the viral targets and positive for the human gene, a pattern that showed that extraction had worked and had introduced no viral target. An internal control is amplified from every patient's specimen (Chapter 16). When it shares the well with the target, as in a multiplex test, it also shows that amplification worked in that very well. In the CDC test it was the human gene RNase P, present in the cells that any good swab collects. All the CDC probes used the same dye, so RNase P ran in a well of its own (Chapter 22). It showed that the specimen contained human material, that extraction recovered it and that amplification worked in the control's own well.
Under the CDC's rules, the controls were read together. A run was valid only if the positive control was positive for all three targets, N1, N2 and RNase P, the no-template control showed no amplification at all and the human specimen control behaved as expected. Within a valid run, a specimen with both viral targets detected was positive, whether or not RNase P was detected. A specimen with neither viral target, but with RNase P detected, was negative. A specimen with only one viral target was inconclusive and was retested. A specimen with nothing detected, not even RNase P, was invalid, and it was to be extracted again or collected again.
The limit of detection
The most important performance characteristic of a detection test is how little target it can detect, its limit of detection (Chapter 6). Laboratory medicine defines this limit carefully. The Clinical and Laboratory Standards Institute, CLSI, sets out the definitions in its guideline EP17-A2, whose second edition it published in 2012 and reaffirmed in 2017. The guideline distinguishes the limit of blank from the limit of detection. The limit of blank is the highest result expected from a sample with no analyte, the substance the test detects, and the limit of detection is the lowest amount that can be reliably told apart from the limit of blank. For qualitative molecular tests the limit of detection is usually defined by hit rate: the lowest concentration at which at least 95 % of replicates are positive. In practice, developers test 20 replicates and require at least 19 positive.
The CDC established the test's limit by spiking a synthetic copy of the N gene's RNA into a suspension of human cells in transport medium, then extracting it and testing replicates at a series of concentrations. The test's instructions give the limit as about 3 RNA copies per microliter of the spiked sample with one extraction system, and 1 copy per microliter with another, while the CDC's paper describing the test gives it as about 5 RNA copies per reaction. The paper's figure comes from a separate experiment: synthetic RNA, diluted in buffer and added directly to the reaction, was detected in all 24 replicates at 5 copies. Five copies per reaction sits just above the sampling floor of Chapter 6, the best that any PCR can do. Below a test's limit, detection does not stop abruptly but becomes hit-or-miss, and hit-or-miss detection is where inconclusive results come from: one target is detected and the other is not.
Inclusivity, exclusivity and interference
A test must detect every version of its target and nothing else. Inclusivity is the first half of that demand: the primers and probe must match all known variants of the organism. During the pandemic, the FDA set out what it expected in a template for emergency submissions. Developers were to show that all published SARS-CoV-2 sequences would be detected, by aligning their oligonucleotides against those sequences, and they were to keep checking as new variants appeared.
Exclusivity, or cross-reactivity, is the second half: the test must not react with related organisms, or with others likely to be in the same specimen. The same template called for laboratory testing wherever an oligonucleotide showed a similarity above 80 % to another organism, and it listed organisms to test at high concentrations. The CDC's scientists reported no cross-reaction with the four common human coronaviruses or with MERS coronavirus. The test's third target, N3, had been designed to detect a broader group of related coronaviruses, and it did react with the 2003 SARS virus, as intended. Interference testing checks that substances likely to be in specimens, such as blood or mucus, do not change the results, and another CLSI document, EP07, guides it.
Clinical performance, and why prevalence changes everything
Analytical performance describes the test on prepared samples. Clinical performance describes it on patients. Against a reference standard, a method that establishes who truly has the condition, a test's sensitivity is the fraction of true cases it calls positive, and its specificity the fraction of non-cases it calls negative. Often no perfect reference exists, and the new test is compared with another test instead.
In its statistical guidance of 2007, the FDA recommends that such comparisons be reported as positive percent agreement and negative percent agreement, not as sensitivity and specificity. These are the shares of the comparator's positive and negative results that the new test matches, and the reason for the different names is that agreement with an imperfect comparator is not accuracy. The FDA also recommends that each figure be reported with a 95 % confidence interval. In its emergency template for SARS-CoV-2 molecular tests, the FDA asked for at least 30 positive and 30 negative clinical specimens, with agreement of at least 95 % in each.
A clinician, though, needs the probability that a positive result is right, which is called the positive predictive value. It depends on how common the condition is among the people tested, its prevalence.
These figures are illustrative, computed for this guide. A test with 95 % sensitivity and 99.5 % specificity is used on 10,000 people. Where 1 % are infected, 100 people are infected, and 95 % of them, 95, test positive. The other 9,900 are uninfected, and 0.5 % of them, about 50, test falsely positive. Of about 145 positives, only 95 are real, so the positive predictive value is 95 out of 145, about 66 %. Where 20 % are infected, 2,000 people are infected and 1,900 test positive. Of the 8,000 uninfected, 0.5 %, or 40, test falsely positive. Of 1,940 positives, 1,900 are real, about 98 %. The test did not change. The population did, and the share of positives that are right rose from about two in three to about 98 in 100.
Where false results come from
Every false result has a mechanism, and each mechanism has a defense described somewhere in this guide.
False positives come from contamination, by synthetic material made in the same place as the reagents or by old product (Chapter 20). They also come from oligonucleotides that react with each other in every tube, as the CDC's N3 oligonucleotides did, and from cross-reaction with a related organism (Chapter 5). Mix-ups of specimens or wells are a further source. The no-template control catches some of these, and design and laboratory practice must prevent the rest.
False negatives come from a poor specimen: a swab that collected little, or one taken too early or too late in the infection. They also come from a target concentration below the limit of detection, and from losses in extraction, degraded RNA and inhibitors (Chapter 20). A change in the organism's sequence under a primer or probe causes them too, which is why the FDA's special controls for SARS-CoV-2 multi-target tests require ongoing monitoring of variants. Instrument faults that leave wells too cool, or that misread them, are a further source (Chapters 10 and 19). The internal control catches several of these by turning them into invalid results.
A negative PCR result means that the target was not detected in the material that reached the reaction, above the test's limit, with a valid internal control. It does not exclude infection, because the swab may have missed the virus, or the specimen may have been taken before the virus multiplied or after it cleared. Instructions for use state this limitation, and a test's design can reduce such misses but cannot remove them.
Laboratories that run such tests work to their own standards. The international standard for medical laboratories is ISO 15189:2022, and ISO 17822:2020 is a quality practice guide written specifically for nucleic acid amplification tests for microbial pathogens. CLSI documents cover the validation of multiplex molecular tests and the setting up of molecular testing environments.
The evidence that makes a result believable has five parts: controls, a limit of detection established by hit rate, inclusivity and exclusivity checks by computer and in the laboratory, interference testing, and clinical agreement with confidence intervals. Plan them from the intended use, run them on the complete test system, and keep monitoring inclusivity for as long as the target keeps evolving.
- Each control answers one question: positive control for reagents and instrument, no-template control for contamination, extraction control for extraction, internal control for the specimen and, when it shares the target's well, for that well.
- The limit of detection of a qualitative test is the lowest concentration detected at least 95 % of the time, and the CDC's paper detected about 5 copies per reaction, near the sampling floor, in an experiment separate from the instructions' limits.
- Inclusivity, exclusivity and interference testing establish what the test detects, and clinical agreement is reported as sensitivity and specificity only against a reference standard.
- Predictive value depends on prevalence: in the illustrative example, the same test gets about 66 % of its positives right at 1 % prevalence and about 98 % at 20 %.
One test at national scale.
+ The questionWhat happened when one PCR test had to be designed, made and shipped across a country in weeks?
Weeks, not years
The first genome sequence of the new coronavirus was posted publicly on 10 or 11 January 2020, on behalf of a consortium led by Yong-Zhen Zhang of Fudan University. Which of the two dates applies depends on the time zone. Everything in this guide follows from what happened next: a sequence is all a PCR test needs to begin. Within days, a group at the Charité hospital in Berlin circulated a protocol, and on 23 January the group published a workflow in the journal Eurosurveillance with targets in three viral genes. The Berlin group had designed the targets from the sequence and from related coronaviruses, because no samples of the new virus were yet available to it. One of the paper's authors, Olfert Landt, was the chief executive of TIB Molbiol, a company that sells PCR assays and by November 2023 was a Roche subsidiary (Chapter 24).
The US Centers for Disease Control and Prevention designed its own test. It chose two targets in the N gene, called N1 and N2, and designed both to be specific to the new virus. A third target, N3, was designed to detect a broader group of related coronaviruses, and a fourth reaction detected the human gene RNase P as a control.
On 4 February 2020 the Secretary of Health and Human Services declared that circumstances justified the emergency authorization of diagnostic tests. The same day, the FDA issued an Emergency Use Authorization, or EUA, to the CDC for its panel. An EUA allows a test that has not been cleared or approved to be used while such a declaration is in force. Shipping began the next day. About 200 kits were to go to US laboratories at first, and a similar number to selected laboratories abroad, each kit enough for roughly 700 to 800 specimens. The laboratories qualified to order the kits numbered 115 in the US and 191 abroad.
About 200 kits were to go to US laboratories at first, each enough for roughly 700 to 800 specimens. At the low end, 200 kits × 700 specimens = 140,000 specimens. At the high end, 200 kits × 800 specimens = 160,000 specimens. In the first kits, each specimen took one well for each of the panel's four reactions, N1, N2, N3 and RNase P. The specimens alone therefore needed between 140,000 × 4 = 560,000 and 160,000 × 4 = 640,000 reactions, before the controls in every run. The first US kits were enough for roughly 140,000 to 160,000 specimens, which meant more than half a million reactions.
The test as designed
The panel used everything earlier chapters describe. It was a one-step real-time RT-PCR, in which reverse transcription and real-time PCR run in the same tube (Chapter 20). Each reaction held 20 microliters: 15 microliters of mix and 5 of extract. Each target had a hydrolysis probe carrying the dye FAM and a quencher (Chapter 12). All the probes used the same dye, so each specimen needed a separate well for each target. As designed, each specimen therefore took four wells, for N1, N2, N3 and RNase P. Once N3 was dropped, in February and March 2020, each specimen took three. The profile was a reverse-transcription step, a polymerase activation step and 45 cycles of 3 seconds at 95 °C and 30 seconds at 55 °C, with fluorescence read at 55 °C (Chapter 4).
The panel ran on one named instrument, with one version of its software, in its standard mode and with the passive reference set to none (Chapter 17). A target was detected if it crossed the threshold before cycle 40.00 (Chapter 13). Controls in every run and the internal control in every specimen governed the calls (Chapter 21). The panel's instructions gave the limit of detection as about 3 RNA copies per microliter of spiked sample with one extraction system and 1 with another, and the CDC's paper reported about five copies per reaction from a separate experiment (Chapter 21). It was authorized for laboratories certified for high-complexity testing, and the CDC estimated that the process took about four hours from the start of sample processing to a result.
What went wrong
Within days of the first shipments, public health laboratories reported problems: no-template controls were giving signals, mainly for N3 and to a lesser extent N1, and runs had to be called inconclusive. The CDC's own scientists later analyzed the first lot of kits and found two separate faults. They published the analysis in the journal PLoS ONE in December 2021.
The first was contamination. About 2 % of the no-template controls for N1 from the initial lot showed false reactivity. Sequencing traced the false signal to a synthetic template that the CDC had made at about the same time as the kits, and the template had reached the bulk reagent during quality assessment. The problem did not recur in later lots.
The second was design. The 3′ end of the N3 probe could pair with the 3′ end of the N3 reverse primer. In a tube with no target, the two paired, and the CDC's sequencing found duplex and larger products containing the N3 probe sequence, which gave fluorescence with no template (Chapter 5). False signals appeared in about 97 % of the N3 no-template controls made with the authorized kits' reagents, against 0.5 to 2 % with oligonucleotides from earlier batches and from commercial suppliers. The false signals became more frequent as the reagents aged. The flaw was present in all three sources of the oligonucleotides, but how often it showed depended on the batch. The authors concluded that flaws in both the assay's design and the handling of bulk material caused the first lot's problems.
In October 2023 the Inspector General of the Department of Health and Human Services reported on the circumstances that allowed these faults. The Inspector General found inadequate control of documents, about four different versions of quality-control procedures in use, a lack of separation of duties, viral material and test reagents made in the same core laboratory, and weak oversight. The Inspector General made six recommendations, and all of them had been closed as implemented by January 2025. In November 2020 NPR reported that, according to a CDC internal review, a final quality-control test had indicated that the kit could fail about a third of the time, and that the kit was released anyway. The Inspector General also found that the CDC had a viable test within about two months of the genome's publication.
About two months from the sequence to a working, accurate test is fast by any historical standard. The faults in the first kits were not caused by speed alone. They came from gaps that would have mattered at any speed: an oligonucleotide interaction that development testing did not catch, and a manufacturing environment that let a synthetic target reach the reagents. According to the CDC's internal review as NPR reported it, a third gap was a release that went ahead despite a failing quality-control result. Speed shortened the time in which those gaps could have been found and acted on.
The fix
The repair came quickly once the causes were understood. On 26 February the CDC validated a protocol that left out the third target and used only N1 and N2. By 28 February it was distributing a working kit to public health laboratories. On 29 February the FDA issued a policy that allowed laboratories certified for high-complexity testing to use tests they had validated themselves while their requests for authorization were reviewed, so other laboratories could bring their own tests into service. On 15 March the FDA reissued the CDC's authorization without N3, and it accepted probes with an additional internal quencher and reagents made by commercial suppliers. In their paper describing the test, the CDC's scientists reported that testing with N1 and N2 alone agreed completely with the original three-target interpretation.
Further revisions followed as the test was used: three more master mixes on 30 March 2020, more extraction systems in June and July, pooling of up to four specimens in December 2020, and later home-collected nasal swabs. In pooling, several specimens are combined and tested together as one sample. On 2 July 2020 the CDC received authorization for a multiplex test that detected influenza A, influenza B and SARS-CoV-2, with the human control, in one well with four dyes. In their 2021 analysis, the CDC's scientists noted that the CDC had used a stricter review process for that test, with outside experts and a pilot in three public health laboratories.
Afterward
In July 2021 the CDC told laboratories that it would withdraw its request to the FDA for emergency use authorization of its original panel after 31 December 2021, and it encouraged them to move to a multiplex test that could also detect influenza. In August the CDC clarified that there were no concerns about the panel's performance and that its primer and probe designs would remain public. In August 2026 the FDA still listed the panel among authorized tests, with the most recent letter for it dated 7 March 2023. On 29 June 2026 the Secretary of Health and Human Services terminated the 2020 declaration that justified emergency authorization of COVID-19 diagnostic tests. The termination takes effect on 26 December 2026, and from that date the remaining COVID-19 test authorizations cease to be in effect.
What the case shows
Nothing in the first weeks went wrong with PCR's chemistry. The cycle copied, the probes reported, the instrument heated and read. What failed was the system around the chemistry. The design check of how the oligonucleotides interacted with each other failed (Chapter 5). So did the separation of the work that made synthetic targets from the work that made reagents (Chapter 20). Quality control at release was not strong enough to stop a faulty lot. Even the conditions of the emergency authorization concerned those layers: the FDA waived most of its manufacturing requirements for the kit but kept those for acceptance activities, for nonconforming product and for statistical techniques, the parts that decide whether a lot is fit to ship.
The CDC test also shows how much a test system includes (Chapter 17). The authorization named one instrument and particular reagents and extraction methods, so laboratories with other equipment could not simply run the test, and each extension of the list took data and a revised letter. Speed came from the flexibility of the emergency route. Safety had to come from the same design controls, risk management and quality system that apply to any test, applied under pressure (Chapters 16 and 19).
For any team that may need to build a test quickly, the lessons of February 2020 concern the steps after design. Check every oligonucleotide pair for interaction, by computer and in no-template reactions at scale. Separate synthetic-template work physically from reagent manufacture. Define lot-release criteria before the first lot is made, with enough no-template controls per lot to see a failure rate of a few percent.
- The CDC designed a one-step RT-PCR test with three N-gene targets and an RNase P control, and the test was authorized on 4 February 2020 and shipped from 5 February.
- The first lot failed for two reasons: contamination of N1 reagent by a synthetic template, and an N3 design in which the probe's and reverse primer's 3′ ends paired, giving false signal in about 97 % of the N3 no-template controls made with the kits' reagents.
- A working N1 and N2 kit was distributed by 28 February, the authorization was reissued without N3 on 15 March, and later revisions added more mixes, extraction systems, pooling and specimen types.
- The reactions copied what their primers started, as every PCR does, while the design check, the manufacturing separation and the quality control at release failed, and naming one instrument in the authorization limited which laboratories could run the test.
+ Part VII · What comes next
The cost of the cycle, and what removes it.
Abbott states that its point-of-care COVID-19 test, ID NOW, reports results in 13 minutes or less, and the test does so without thermal cycling, by amplifying its target at a single temperature. PCR's costs are time, a laboratory and an instrument that can move heat quickly and evenly. This part looks at three kinds of method that cut those costs: faster cycling, amplification at a single temperature, and enzymes that detect sequences directly. It asks what each gives up, and it ends by setting out what remains unsolved.
Faster, smaller, simpler.
+ The questionIf PCR's costs are time, a laboratory and an instrument, what do faster and simpler methods give up to remove them?
Faster cycling
In 2015 Carl Wittwer's group amplified short human DNA targets through 35 cycles in 15 to 60 seconds. The cycle takes time because heat takes time to move (Chapters 9 and 10). It shortens when the volume and the distance that heat must travel shrink, and when higher concentrations let each step finish in less time. In the group's method, which it called extreme PCR, the primer and polymerase concentrations were 10 to 20 times above normal, the samples were 1 to 5 microliters and the cycles lasted 0.4 to 2 seconds. The targets were 45 to 102 base pairs long, and they amplified at efficiencies of about 92 to 96 %.
In a 2015 demonstration of photonic PCR, blue light heated a thin gold film under the wells, and the device completed 30 cycles in 5 minutes. A 2019 commentary described the pairing of extreme PCR with rapid melting analysis as a step toward molecular diagnosis while the patient waits.
What faster cycling gives up is visible in the figures: short targets, small volumes and much more reagent per reaction. A smaller volume holds fewer target molecules at a given specimen concentration, so the reaction works closer to the sampling floor of Chapter 6, and high reagent concentrations raise the cost and the risk of side reactions. Fast cycling is also only as good as the temperature control that delivers it. A three-second hold achieves nothing if the liquid never reaches the set temperature.
The 2015 photonic device ran 30 cycles between 55 and 95 °C in 5 minutes, and its authors reported heating at about 12.8 °C per second and cooling at about 6.6 °C per second. Five minutes is 300 seconds, so each of the 30 cycles had 10 seconds. In each cycle the temperature rose and fell by 40 °C. Heating took 40 divided by 12.8, about 3.1 seconds, and cooling took 40 divided by 6.6, about 6.1 seconds, so changing temperature took about 9.2 seconds. That leaves under 1 second of each 10-second cycle for the holds, assuming the rates held across each swing and the 5 minutes went to the cycles alone. At these rates, changing temperature filled more than nine-tenths of the run, and the slower cooling alone filled about three-fifths of it.
Amplifying at one temperature
The other route removes the cycle altogether. Isothermal amplification methods copy nucleic acid at a single temperature, using enzymes that separate strands as they copy instead of relying on heat to separate them. One of them is loop-mediated isothermal amplification, or LAMP, published in 2000 by Tsugunori Notomi and colleagues, most of them staff of the Japanese company Eiken Chemical. LAMP runs at about 65 °C with a polymerase that displaces the strand ahead of it as it copies, and with four primers that recognize six distinct regions of the target. The primers are designed so that the products fold into loops that serve as new starting points, and in the authors' experiments the reaction accumulated about a billion copies in under an hour.
Recombinase polymerase amplification, or RPA, was published in 2006 by a team mostly from ASM Scientific, a company that declared a financial interest in the method, and it pushes primers into double-stranded DNA at 37 °C with proteins borrowed from the machinery that cells use to repair and recombine DNA. In that 2006 work, RPA detected fewer than ten copies of genomic DNA in under 30 minutes. Other isothermal schemes use enzymes that nick, or cut, one strand of the DNA to start each round of copying. ID NOW, the test at the opening of this part and in Chapter 18, amplifies its target isothermally, and Abbott states that it gives results in 13 minutes or less.
Removing the cycle removes the instrument's hardest job: the instrument needs only to hold one temperature, which a simple heater can do, so isothermal tests suit small, cheap, battery-powered readers. The price is paid in the chemistry. Many primers in one tube give many chances for primers to interact, and isothermal reactions are prone to amplifying something even without a target. Without cycles, there is no Cq built on doubling per cycle, so quantification is harder. And the design of four to six primers per target constrains where on a genome a test can be placed.
A method that reaches a result in minutes rather than an hour usually does so with a different chemistry, a smaller volume, fewer cycles or a different target. Its limit of detection, specificity and failure modes must be established on their own, not inferred from the slower test it replaces.
Enzymes that recognize a sequence
A third approach, CRISPR detection, adds sequence recognition by the bacterial immune enzymes known as CRISPR systems. Some of these enzymes are guided by a short RNA that matches a target, and once they bind their target they become indiscriminate cutters, cutting any nearby single-stranded nucleic acid. When the enzymes are mixed with a reporter molecule that fluoresces once it is cut, the recognition of one sequence becomes a signal.
Published in 2017, the method called SHERLOCK used the enzyme Cas13 after a pre-amplification by RPA, and its developers reported sensitivity down to attomolar concentrations, on the order of one molecule per microliter, with the ability to distinguish single-base differences. DETECTR, published in 2018, used Cas12a and distinguished two closely related types of human papillomavirus. In May 2020 the FDA issued an emergency authorization for a CRISPR-based SARS-CoV-2 test, the Sherlock CRISPR SARS-CoV-2 Kit, and its maker, Sherlock Biosciences, called it "the very first FDA-authorized use of CRISPR technology".
CRISPR detection recognizes a sequence but does not amplify it. On its own it is not sensitive enough for a few copies, so it is usually paired with an isothermal amplification step, and it inherits that step's strengths and weaknesses. What it adds is a second check of sequence identity, like a probe's, in a format that a simple fluorescence reader or a paper strip can read.
Where PCR stays
None of these methods has displaced PCR, and the reasons are the subject of this guide. PCR's chemistry is understood in detail, and its primers are simple to design. Its exponential phase gives a Cq that can be turned into a quantity, its digital form counts molecules, and its instruments, assays, controls and regulatory routes are mature. Most laboratory molecular diagnostics run on real-time PCR, and sample-to-answer cartridges have brought it close to the patient. Digital PCR measures rare variants in blood, and PCR prepares the DNA for most sequencing. The newer methods find their places where time, cost or portability matter more than quantification and the breadth of established tests.
What remains unsolved
Several problems are common to every method in this chapter, and none is solved. Sample preparation remains the slowest, least standardized and least integrated step outside the cartridges of Chapter 18: amplification takes minutes, but getting clean nucleic acid out of a specimen takes skill or an expensive disposable. Comparability remains elusive, because a Ct from one test cannot be compared with a Ct from another (Chapter 13). The field's move toward vendor-neutral raw data and efficiency-corrected quantities is recent. Evolving targets are a lasting problem: every test that detects a virus must be checked against each new variant for as long as the test is used. And cost per result still separates what is technically possible from what a laboratory, a clinic or a health system can provide at scale.
This guide asked how heating and cooling a sealed tube can find one stretch of DNA among billions, copy it a billion times and count how much of it the sample held. The chemistry answers the first two parts. The last part depends on the engineering of the instrument as a system and on the evidence built around the test, and so does the trust a laboratory places in the answer. Most of the remaining work lies in those two.
Before choosing a faster or simpler method, name which of PCR's costs matters for the intended use: time to result, the need for a laboratory, the price of the instrument, the price per test or the skill to run it. Each alternative removes some of these and adds others, and the right choice follows from the intended use, as every other decision in this guide does.
- Extreme PCR has completed amplification in a minute or less and photonic PCR in five minutes, at the cost of short targets, small volumes and high reagent concentrations.
- Isothermal methods such as LAMP at about 65 °C and RPA at 37 °C remove thermal cycling and suit simple readers, but they are harder to design, more prone to non-specific amplification and harder to quantify.
- CRISPR enzymes add sequence recognition, usually after an amplification step, and in May 2020 a CRISPR-based SARS-CoV-2 test became, by its maker's account, the first FDA-authorized use of the technology.
- PCR remains central, while sample preparation, comparability of results, evolving targets and cost per result remain unsolved.
+ Part VIII · The field in 2026
Who makes the instruments, reagents and tests.
Between 9 February and 17 August 2026, BD's diagnostics business combined with Waters, two investment firms took Hologic private, and Molbio Diagnostics listed its shares on India's stock exchanges. The cyclers, cartridges and reagents of the earlier parts are made and sold by companies like these. This part, one chapter dated October 2026, describes who makes PCR products, layer by layer: open instruments and digital PCR, central-laboratory systems, cartridges, home and point-of-care tests, and reagents, then the makers in India and the businesses that changed hands from 2023 to 2026. It gives each product's record with its regulator where there is one, and it ranks none of them.
The makers, layer by layer.
+ The questionWhich companies make the instruments, cartridges and reagents that PCR runs on in 2026, and what does the regulator's record show for each?
Open instruments, real-time and digital
The CDC's test of February 2020 ran on the Applied Biosystems 7500 Fast Dx, which FDA cleared in 2008 through a 510(k), the US route for a device shown to be substantially equivalent to one already on the market (Chapter 25). Thermo Fisher now sells the Applied Biosystems line. When FDA reissued the CDC's authorization on 15 March 2020, it accepted reagents made by commercial suppliers. One PCR result therefore rests on several makers, of the instrument, the reagents and the test, and each works in a different layer of the field. This chapter takes the layers in turn, as of October 2026. It gives the record of a regulator or public body for each product where one exists, names the source of every other status, and ranks no product.
The first layer holds the open platforms: real-time instruments that run any assay written for them, a kit maker's or a laboratory's own. Thermo Fisher's QuantStudio, Bio-Rad's CFX, Roche's LightCycler and Qiagen's Rotor-Gene Q are the families in this chapter. Several diagnostic versions carry "Dx" or "MDx" in their names, and the older ones hold 510(k)s, from the LightCycler 1.2 in 2003 to the QuantStudio Dx in 2013. The newer QuantStudio 5 Dx, QuantStudio 7 Pro Dx, CFX Opus Dx and CFX96 Dx appear in FDA's listing, the database of the devices that makers sell, with no 510(k) number. As of October 2026, FDA's regulation for this type of instrument exempts it from the 510(k), within limits, and Roche stated in November 2023 that its new LightCycler PRO would be exempt as well.
For the digital PCR of Chapter 14, a search of FDA's 510(k) records by device name finds three tests, and all three monitor one marker, BCR-ABL1, the fusion gene of chronic myeloid leukemia, formed by an exchange between chromosomes 9 and 22. Bio-Rad's QXDx BCR-ABL %IS Kit for its QXDx droplet system was cleared on 13 February 2019, and kits from Suzhou Sniper Medical Technologies and Suzhou Precigenome followed on 5 September 2023 and 2 July 2026. Qiagen stated at the launch of its nanoplate instrument QIAcuityDx in 2024 that the instrument was certified under the IVDR, the EU's regulation for diagnostic devices, and exempt from the 510(k). Under the IVDR a PCR instrument is class A, so its maker declares it conformant itself, without a notified body, the designated assessor that Chapter 26 describes, and the word certified is Qiagen's. Stilla Technologies had announced the end of its naica system, and Bio-Rad, which bought Stilla in 2025, is relaunching Stilla's nio as its own QX700 series. Bio-Rad labels the QX700 E research use only, not for diagnosing patients.
Closed systems, from the central laboratory to the cartridge
Central-laboratory systems automate extraction, PCR and reading for many specimens at once. In FDA's 510(k) records, Roche's cobas 5800, 6800 and 8800 systems hold tests cleared from March 2018 to 8 July 2026, Hologic's Panther Fusion from September 2017 to September 2025, and Abbott's Alinity m from April 2022 to September 2025. BD's molecular systems, BD MAX and BD COR, belong to the diagnostics business that combined with Waters in February 2026, and BD states that it remains their legal manufacturer during the transition.
Cartridges, the sealed disposables of Chapter 18, carry the same chemistry, and some are syndromic panels, which test one specimen for the many pathogens that can cause the same symptoms. Several of Cepheid's Xpert Xpress tests, Roche's cobas liat tests and the BioFire Spotfire panels of bioMérieux were cleared as Dual Track submissions. Each was a 510(k) filed together with an application for a waiver under CLIA, the US law on clinical laboratories, whose least demanding category holds simple tests with a low risk of error (Chapter 25). Cepheid's Dual Track records run from December 2015 to a hemorrhagic fever test of 7 August 2026, the cobas liat's from July 2016, first in the name of IQuum, which Roche acquired, and the Spotfire's from February 2023. Qiagen's QIAstat-Dx panels hold 510(k)s from May 2019 to September 2026 with no Dual Track record: FDA's records list them as traditional 510(k)s, filed without a waiver application.
Home and point of care
Of the three home molecular tests in this chapter, one is PCR. Visby Medical's Women's Sexual Health Test is a single-use device that amplifies DNA by thermocycling from a self-collected swab and reports chlamydia, gonorrhea and trichomoniasis in under 30 minutes. FDA authorized it for over-the-counter home use on 28 March 2025 through the De Novo route, for a new kind of device with nothing equivalent on the market. Visby calls it the first PCR device approved by FDA for home use for any indication, though FDA's record shows the De Novo request granted, and FDA calls the result a marketing authorization, not an approval (Chapter 25). Visby's newest clearance, for an influenza and COVID-19 test, came on 27 May 2026.
The other two copied their targets at one temperature (Chapter 23). Cue Health's 20-minute COVID-19 test received a De Novo authorization for over-the-counter use on 6 June 2023. Cue's bankruptcy filing was reported on 29 May 2024, and FDA revoked Cue's two emergency authorizations, effective 9 October 2024, after Cue said it had stopped making the test. Lucira Health, whose tests used LAMP, filed for bankruptcy on 22 February 2023, and Pfizer bought its assets in April 2023 and later ended the business, citing the cost of the tests. FDA revoked both Lucira by Pfizer authorizations, effective 22 October 2025, because no viable reagents remained in the United States. Both isothermal home tests have left the market, and at the point of care Abbott's ID NOW, the test of Chapters 18 and 23, is isothermal too.
A name or a press release can say "molecular" without saying whether the test cycles. The authorization documents say which: "thermocycling" for Visby's test, "isothermal nucleic acid amplification technology" for Cue's test and for ID NOW. A comparison of PCR products that counts isothermal tests sets two chemistries side by side, with the different failure modes of Chapter 23.
Reagents and oligos
Every layer depends on the makers of polymerases, master mixes and oligonucleotides, or oligos: the short synthetic strands of DNA that serve as primers and probes. Integrated DNA Technologies, IDT, makes oligos as an operating company in Danaher's life sciences segment, and Danaher's diagnostics segment holds Cepheid, so one owner holds both an oligo maker and a cartridge maker. Roche sells assays through its subsidiary TIB Molbiol, and by Roche's count of November 2023 more than 60 of them carried the CE mark for diagnostic use in the EU. New England Biolabs states that it manufactures in a facility in Ipswich, Massachusetts, certified to ISO 13485:2016, the medical device quality standard that FDA's quality regulation now incorporates (Chapter 25). Takara Bio states that it offers manufacturing to good manufacturing practice in Japan, and its US site marks its products for research use only. Chapter 22 showed the limit of what a supplier controls: the CDC's N3 flaw appeared in oligos from all three of the CDC's sources.
India
Molbio Diagnostics makes Truenat, a portable, battery-operated system whose Truelab analyzer runs real-time PCR on a chip and gives results in under an hour. On 2 July 2020 the Indian Council of Medical Research, ICMR, and Molbio announced that WHO had endorsed three Truenat assays for the initial diagnosis of tuberculosis and for detecting resistance to the drug rifampicin, after an ICMR validation found them comparable to Cepheid's Xpert MTB/RIF. Such an endorsement is a WHO recommendation that a type of test may be used for a purpose. Prequalification is WHO's separate assessment of a named product, and as of October 2026 WHO lists Truenat MTB Plus as under assessment for it.
During the pandemic, ICMR had most commercial RT-PCR kits evaluated at centers it designated before India's central regulator, the Central Drugs Standard Control Organisation, CDSCO, approved them (Chapter 26). Its summary of 17 February 2022 counted 523 kits evaluated and listed 211 as satisfactory, each only for the batch numbers given. The Indian makers on the list include Mylab, Huwel, Genes2Me, Meril, CoSara and Trivitron, and Truenat is not on it.
ICMR's summary gives 523 kits evaluated and 211 listed as satisfactory. The share is 211 ÷ 523 = 0.403, or about 40 %. The remaining 523 − 211 = 312 kits, about 60 %, are not on the list. The count is of evaluations at ICMR's centers, not of CDSCO approvals. In words: of the RT-PCR kits put through ICMR's evaluation by February 2022, fewer than half were listed as satisfactory.
Two of the listed makers were each described as first in March 2020. The US company Co-Diagnostics stated on 19 March that its joint venture CoSara was the first Indian company licensed by CDSCO to manufacture RT-PCR kits for COVID-19, and on 24 March a trade report carried Mylab's claim that its kit was the first to receive commercial approval. The claims may describe different steps, a license to manufacture and an approval to sell. In February 2026 Co-Diagnostics announced that CoSara had received a license on Form MD-5 to manufacture its CoSara PCR Pro, a point-of-care real-time PCR instrument, in Gujarat. CDSCO's list puts a PCR analyzer in class A, the lowest of India's four risk classes, and for class A a state licensing authority issues the MD-5 license (Chapter 26). The instrument's kits for tuberculosis and for human papillomavirus still needed their own clearance.
Changes of hands, 2023–2026
On 6 June 2024 Qiagen announced that it would discontinue its NeuMoDx integrated PCR systems, citing the market after the pandemic, with a restructuring charge it put at about $400 million. On 29 July 2024 Roche completed its purchase of LumiraDx's point-of-care technology, an immunoassay and clinical chemistry platform that is not PCR. Bio-Rad gives 30 June 2025 as the date it finalized its purchase of Stilla Technologies, and trade reports of the completion followed on 7 and 8 July. BD's Biosciences and Diagnostic Solutions businesses were spun off and combined with Waters on 9 February 2026. Blackstone and TPG completed their purchase of Hologic on 7 April 2026, taking it private, and Molbio listed its shares in India on 17 August 2026.
New tests kept arriving: FDA cleared at least eight PCR tests from Roche, Visby, Suzhou Precigenome, Cepheid and Qiagen between 15 May and 24 September 2026. The words in these records, cleared, listed, authorized, waived, licensed and endorsed, name different acts, and Chapters 25 and 26 describe the systems behind them.
Before choosing an instrument, a cartridge platform or a reagent supplier, look up its status in the record of every market it will serve, check whether it is labeled for research use or for diagnosis, and treat a status that only the company states as a claim to verify. Then plan for the supplier changing: four product lines in this chapter had ended by 2026 and five businesses changed hands between 2023 and 2026, so a design that depends on one maker needs a second qualified source or a plan for its withdrawal.
- The newer diagnostic open instruments are listed with FDA without 510(k)s, and the three digital PCR tests found in FDA's 510(k) records all monitor BCR-ABL1.
- Cartridge tests divide in FDA's records between Dual Track clearances, with a CLIA waiver application, and 510(k)s filed without one, which FDA's records call traditional.
- Of the three home molecular tests in this chapter only Visby's is PCR, and both isothermal ones have left the market.
- India's records come from ICMR, WHO and makers' statements of CDSCO licenses, and four product lines in this chapter had ended by 2026 and five businesses changed hands from 2023 to 2026.
+ Part IX · Quality and regulation, as of October 2026
Deciding when a count can be trusted.
On 29 June 2026 the Secretary of Health and Human Services terminated the declaration that had justified emergency authorization of COVID-19 diagnostic tests since 4 February 2020, with effect from 26 December 2026. In a laboratory advisory of 14 October 2026, the CDC told laboratories what follows: the FDA does not intend to object to continued distribution of a test whose marketing submission it has accepted by 26 December 2026, and laboratories may go on using tests already distributed until they expire. The earlier chapters showed why each requirement on a test exists, from the limit of detection to the controls in every run and the instrument named in every authorization. This part maps those requirements onto the rules of three jurisdictions, giving the reason for each rule before its clause, with every status dated October 2026: the United States in Chapter 25, and the European Union and India in Chapter 26.
The device and the laboratory.
+ The questionWhat must a PCR instrument and test show before US laboratories may use them on patients, and who decides?
Three questions behind every route
Within days of receiving the CDC's first kits in February 2020, some public health laboratories could not verify them, and the emergency authorization under which the kits had shipped already said what had to happen next. Each such laboratory had to notify the CDC, and the CDC had to notify the FDA (Chapter 22). That condition is a small instance of how US regulation acts on a test. It attaches obligations to a named product for a stated use, and it rests on evidence that answers three questions.
The rules differ between jurisdictions, but the evidence they ask for answers the same three questions, in order, and the European Union's regulation for diagnostic devices names them precisely. Scientific validity asks whether the thing measured, the analyte, is associated with the clinical condition: for a SARS-CoV-2 test, whether the virus's RNA in a respiratory specimen is associated with infection. Analytical performance asks whether the device detects or measures the analyte correctly, on the complete test system, instrument included. Clinical performance asks whether the results correspond to the condition in the people the test is meant for. Chapter 21 set out what each of the last two contains.
A fourth question, whether using the test improves outcomes for patients, is its clinical utility. It matters to doctors and to those who pay for tests, but market authorization generally does not require it. In the United States, two bodies of law divide the work. The FDA regulates the instrument and the test as medical devices, and a separate law, the Clinical Laboratory Improvement Amendments, or CLIA, regulates the laboratories that run them. Every status in this chapter is stated as of October 2026.
Classes and routes for the instrument and the test
The FDA sorts devices into three risk classes, from class I, the lowest, to class III, the highest, and by product type. Each type has a classification regulation that names its class and, for class II, its special controls: requirements that every device of that type must meet. A device then reaches the market by one of a few routes. A 510(k), or premarket notification, shows that the device is substantially equivalent to one already legally marketed, its predicate, and the FDA clears it. A device of a new type with no predicate, and of low to moderate risk, can instead go through a De Novo request, which the FDA grants. The grant creates a new classification, and the granted device can then serve as the predicate for later 510(k)s.
Real-time PCR instruments are classified under 21 CFR 862.2570, instrumentation for clinical multiplex test systems, as class II devices with special controls. A final order of 30 December 2019 exempted them from premarket notification, subject to the general limits on exemptions. The instrument named in the CDC's test had been cleared through a 510(k) in September 2008, before that exemption.
The exemption removes the premarket notification and nothing else. The instrument remains a class II device, its product type is not exempt from the manufacturing requirements, and the quality management system regulation described in this chapter's last section requires design controls for class II devices. The exemption also covers the instrument alone: a test sold for use on it still needs its own authorization, for its intended use with specified instruments (Chapter 17).
Tests follow the same logic. For tests that detect SARS-CoV-2 together with other respiratory pathogens, the FDA created a classification, 21 CFR 866.3981, through the De Novo route. It granted a De Novo request for BioFire Diagnostics' Respiratory Panel 2.1 on 17 March 2021, and the trade press reported it as the first SARS-CoV-2 test authorized through a premarket route rather than an emergency authorization. The classification was codified on 16 August 2024. Its special controls turn the second and third questions into required studies of the limit of detection, cross-reactivity and clinical performance. They also require labeling stating that a negative result does not rule out infection and that sequence variants can cause incorrect results, and continuous monitoring of viral variants (Chapter 19). Results suggesting an effect on performance go to the FDA immediately, and evaluations within 48 hours if the agency asks. Later tests of the same type reach the market through 510(k)s that cite earlier ones, as Cepheid's Xpert Xpress CoV-2/Flu/RSV plus did in August 2023.
Emergency use, and how it ends
An Emergency Use Authorization, or EUA (Chapter 22), lets the FDA allow an unapproved product during a declared emergency. Under section 564 of the Federal Food, Drug, and Cosmetic Act, the agency may authorize a test if the disease is serious or life-threatening, the test may be effective, its known and potential benefits outweigh its known and potential risks, and no adequate, approved and available alternative exists. An authorization lasts until the declaration behind it ends or the FDA revokes it, whichever comes first. Ending a declaration requires advance notice in the Federal Register, with reasonable time to dispose of the products. Authorizations granted outside an emergency eventually remove the fourth criterion, the lack of an adequate, approved and available alternative: on the day it granted the De Novo request for the BioFire panel, the FDA revoked that panel's emergency authorization, because an adequate approved alternative now existed.
The declaration for COVID-19 tests was made on 4 February 2020 and amended on 15 March 2023. On 29 June 2026 the Secretary of Health and Human Services, Robert F. Kennedy, Jr., terminated it, with effect from 26 December 2026. HHS judged 180 days a sufficient period for disposition of the products, and its notice observed that more than fifty COVID-19 tests held marketing authorization granted outside an emergency. The FDA states that from 26 December the COVID-19 device authorizations on its list will no longer be in effect.
The end comes with a transition. The FDA's guidance of March 2023 asks a manufacturer that intends to keep distributing an emergency-authorized test to submit a marketing application in time for the agency to accept it before the termination date. The FDA does not intend to object to continued distribution once it has accepted the submission and while it has taken no final action. Under the same guidance, single-use tests distributed before the termination date may be used until the expiry date listed on that day, which cannot be extended. In its laboratory advisory of 14 October 2026, the CDC repeated these points for laboratories. It added that the Centers for Medicare & Medicaid Services lets laboratories keep using such tests, as their authorizations allowed, until the product's expiration date, and it asked laboratories to review the emergency-authorized tests they run and plan their transition.
Laboratories: CLIA, and tests made in one laboratory
CLIA regulates the laboratory, not the product. It sorts tests into three categories by the complexity of performing them: waived, moderate complexity and high complexity. A waived test meets separate criteria of simplicity and low risk of error. The others are scored on seven criteria: knowledge; training and experience; reagent preparation; operational steps; calibration, quality-control and proficiency-testing materials; troubleshooting and maintenance; and interpretation and judgment. The FDA categorizes commercially marketed tests, during premarket review or on request, and a manufacturer can apply separately for a waiver. In the FDA's 510(k) records, a submission that combines a 510(k) with an application for a waiver is marked Dual Track. The CDC's panel was authorized only for laboratories certified for high-complexity testing.
Each of the seven criteria is scored 1, 2 or 3, so totals run from 7, the lowest grade everywhere, to 21. A total of 12 or less is moderate complexity, and anything above is high. A test of middle grade on every criterion scores 7 × 2 = 14, which is high complexity. From the minimum of 7, a test has 12 − 7 = 5 points to spend: a criterion scored 2 spends 1 point, and one scored 3 spends 2. Each criterion above the lowest grade spends at least 1 point, so at most five criteria can rise above it, and five at 2 spend all 5 and leave the other two at 1. Three criteria at 3 would spend 6, one too many, so at most two can score 3. In words: a moderate-complexity test must score the lowest grade on at least two of the seven criteria and the highest on no more than two.
CLIA also covers tests that never pass through the FDA. A laboratory-developed test is a test made and used within a single laboratory. In May 2024 the FDA issued a rule stating that such tests are devices even when the manufacturer is a laboratory, and phasing out its long-standing practice of not enforcing device requirements on them. On 31 March 2025 a federal district court in Texas vacated the rule, holding that such tests are professional services rather than the products the law's device provisions cover. On 19 September 2025 the FDA removed the rule's words from its regulations. As of October 2026, laboratory-developed tests remain subject to CLIA. CLIA's regulations require a laboratory that introduces a test it developed, or modifies one the FDA has cleared or approved, to establish the test's performance specifications before reporting patient results, including, as applicable, accuracy, precision, analytical sensitivity and specificity (Chapter 21), reportable range and reference intervals.
The quality system since 2 February 2026
The design controls of Chapter 16 were already required in the United States under the FDA's earlier quality system regulation. They are now required through the Quality Management System Regulation, or QMSR, which the FDA issued as a final rule on 2 February 2024 and which took effect on 2 February 2026. It incorporates ISO 13485:2016 by reference, with clause 3 of ISO 9000:2015 for its terms, so the design controls, risk management and production controls of a diagnostic manufacturer follow the international standard. The FDA adds its own requirements on unique device identification, complaints and medical device reports, corrections and removals, and traceability. Design controls apply to class II and class III devices and to listed class I devices. A real-time PCR instrument under the exempt classification is a class II device, so its maker carries out design controls without ever filing a 510(k).
Write the intended use first, then find the classification regulation that fits it and read its special controls as a list of studies the evidence must contain. Decide early which laboratories the test is for, because the CLIA category sets how much of the procedure the design must take from the user, and a waiver is a separate application. For an instrument under an exempt classification, plan design controls, the quality system and the labeling as for any class II device. For a test that relies on an emergency authorization, have a marketing submission accepted before the emergency ends.
- Regulators ask in turn whether the analyte is associated with the condition, whether the device measures it correctly on the complete test system, and whether its results match the condition in the intended population.
- In the US, real-time PCR instruments are class II and have been exempt from premarket notification since December 2019, while multi-target SARS-CoV-2 tests are class II under a classification created by a 2021 De Novo, whose special controls require analytical, clinical and variant-monitoring work.
- COVID-19 emergency authorizations for devices end on 26 December 2026, the FDA does not intend to object to a test staying on the market if it has accepted the test's marketing submission by then and taken no final action, and single-use tests already distributed may be used until they expire.
- CLIA sorts tests into waived, moderate and high complexity, laboratory-developed tests remain subject to it as of October 2026 after the 2024 rule was vacated, and since 2 February 2026 the QMSR has incorporated ISO 13485:2016 into US device regulation.
Same evidence, different assessors.
+ The questionThe United States separates the device from the laboratory. How do the European Union and India classify and assess the same instrument and test, and what does each add?
Four classes, and a test that changed class
In March 2025 a laboratory SARS-CoV-2 test in the European Union moved from the highest of four risk classes to the second lowest, although nothing about the test had changed. What had changed was the guidance on how tests for the virus are classified. The EU's regulation on in vitro diagnostic devices, Regulation (EU) 2017/746, or IVDR, has applied since 26 May 2022. It sorts devices into classes A to D by seven rules, and the class decides who assesses a device. Manufacturers self-declare class A devices that are not sterile, while classes B, C and D need a notified body, an organization designated to assess conformity.
Rule 5 puts in class A the instruments that a manufacturer specifically intends for diagnostic procedures, and the classification guidance of the Medical Device Coordination Group, MDCG 2020-16, names a PCR thermocycler among its examples. Rule 1 puts in class D devices that detect a transmissible agent causing a life-threatening disease with a high or suspected high risk of spreading, and SARS-CoV-2 tests were classified there until 2025. The March 2025 revision of the guidance removed SARS-CoV-2 from its examples for Rule 1. It followed advice from an expert panel that the virus no longer posed a life-threatening risk with significant mortality to the general EU population. Professional SARS-CoV-2 tests are now treated as class B under the general Rule 6, and self-tests as class C under Rule 4.
The revision of September 2026 keeps that reading and notes that a change in the epidemiological context could put such tests back in a higher class. The guidance is not legally binding. Notified bodies published a position on 12 December 2025 on replacing the class D certificates of SARS-CoV-2 tests with certificates for class B or C.
Common specifications and reference laboratories
For class D the IVDR adds two layers. Common specifications, set by implementing regulation, add detailed performance requirements for certain class D devices, and the set that has applied since 25 July 2024 includes an annex on SARS-CoV-2. Whether that annex still binds tests now treated as class B had not been clarified as of October 2026. EU reference laboratories verify the performance of class D devices and their compliance with common specifications, and they test batches. The first five began work on 1 October 2024, in areas that include respiratory viruses.
Deadlines, and a proposal to simplify
Regulation (EU) 2024/1860 gives more time to legacy devices, those placed on the market under the previous directive. A device certified under that directive may stay on the market until 31 December 2027. A device that needed no notified body before but needs one now may stay until 31 December 2027 in class D, the end of 2028 in class C and the end of 2029 in class B or as a sterile class A device. Both depend on conditions: no significant change in design or intended purpose, an IVDR quality system by 26 May 2025, and an application to a notified body and a signed agreement with it by set dates. Those dates fell in 2025 for certificate holders and class D, in 2026 for class C, and in 2027 for class B and sterile class A, so as of October 2026 only the last group could still apply.
On 16 December 2025 the European Commission proposed to simplify the device and diagnostic regulations. For diagnostics its changes include replacing the fixed five-year validity of certificates with risk-based review and widening the exemption for tests made within a health institution. Its memorandum proposes no change to the transition end dates. As of October 2026 the proposal had not been adopted, and in the European Parliament it still awaited a committee decision. The news service Agence Europe reported that on 14 October 2026 member states' representatives backed the Irish Presidency's aim of agreeing the Council's negotiating position by 31 December 2026.
India: classes, licenses and evaluation on three batches
India's Medical Devices Rules, 2017, in force since 1 January 2018, sort diagnostic devices into classes A to D from low to high risk. The Central Drugs Standard Control Organisation, or CDSCO, publishes the class of each device type. As of October 2026 its newest list was the update of 25 October 2023. It places a nucleic acid amplification (PCR) analyzer in class A, and COVID-19 RT-PCR and isothermal kits in class C, as it does RNA and DNA extraction kits. Licensing became mandatory for classes A and B from 1 October 2022 and for classes C and D from 1 October 2023. State licensing authorities license manufacturers of class A and B devices, and the central licensing authority licenses those of class C and D.
In October 2022 India exempted non-sterile, non-measuring class A medical devices from licensing. CDSCO's answers to frequent questions on diagnostic devices state that this category does not apply to them, and its list of such devices, published on 31 October 2025, holds 382 entries and no diagnostic device or PCR instrument. A PCR analyzer is class A and still needs a manufacturing license.
India adds a step of its own. A new diagnostic device in class B, C or D needs a performance evaluation on three batches made from different lots of key raw materials, by a laboratory set up or designated by the central government or accredited by the national accreditation board. During the pandemic the Indian Council of Medical Research, or ICMR, organized this check. Its guidelines of 4 June 2020 required the first batch of most RT-PCR kits to be validated at one of 24 designated centers before approval, and Chapter 24 gives the count of kits evaluated.
An undated ICMR protocol for validating commercial RT-PCR kits used 75 positive specimens, spread across low, medium and high Ct values, and 85 negative ones, 10 of them positive for other respiratory viruses. It required a sensitivity of at least 94.7 % and a specificity above 98.8 %. For sensitivity, 71 of 75 is 94.7 % and 70 of 75 is 93.3 %, so a kit could miss at most 4 positives. For specificity, 84 of 85 is 98.82 % and 83 of 85 is 97.6 %, so a kit could give at most 1 false positive. In words: a kit passed only with no more than four missed positives and one false positive, and the ten negatives with other viruses also tested cross-reaction.
ICMR and CDSCO have since published standard evaluation protocols. Drafts were released for comment from 31 December 2024, the first for dengue, chikungunya and Zika tests, and in September 2025 a compendium of 39 protocols followed, among them real-time PCR protocols for human metapneumovirus and for multiplex respiratory panels.
One body of evidence, filed three ways
The same instrument and the same test land in different classes in each system. A PCR instrument is class A in the EU and in India, and class II but exempt from premarket notification in the United States. A laboratory SARS-CoV-2 test is class B in the EU, class C in India and, as part of a multi-target respiratory test, class II in the United States. Routes, documents and assessors differ.
The evidence underneath them barely differs. Scientific validity, analytical performance on the complete test system, clinical performance with confidence intervals, a risk file, software documentation, safety and compatibility test reports for the instrument, and a quality management system to ISO 13485 serve all three. A developer can organize that core by the system's requirements and their verification (Chapters 16 and 19), build it once and present it three ways. Building it three times costs three times as much and risks three inconsistent answers.
Each word names a different act. A 510(k) is cleared, a De Novo request granted, an emergency use authorized and a premarket approval approved. In the EU a notified body issues a certificate for a device of class B, C or D or a sterile class A device, and the CE marking follows. The maker of a non-sterile class A device declares conformity itself. In India a manufacturer is licensed, and a registrar certifies a quality system. A request that has only been filed has none of these statuses, and no consultant or test laboratory approves or certifies a device. Every regulatory claim about a test should name the status, the regulator, the intended use and the date.
The class of an instrument and of a test in each target market decides the route, the assessor, the timeline and much of the evidence, so settle it at the start, beside the intended use. Check the class in each market against current guidance and lists, plan one evidence core for all of them, and check again before each submission, because a class can move with the epidemiology.
The CDC panel, at the end of its emergency
This book opened in February 2020, when public health laboratories verifying the CDC's first kits found that the N3 no-template wells, which held no sample, crossed the threshold. In the CDC's later analysis, 97 % of the N3 no-template controls made with the kit's reagents gave false signals at Ct 34 to 39, inside the cut-off of 40. The 3′ end of the N3 probe could pair with the 3′ end of its reverse primer, so a tube with no target could still make a product (Chapter 5). On 28 February the CDC said that the third component could be left out.
The panel holds the whole answer. Two primers matched one stretch of the viral genome among the human DNA of a swab, and each cycle at most doubled it, so thirty doublings could make a billion copies. The chemistry also copied whatever else its primers started, and in an empty N3 well that was a probe paired with a primer. The instrument did its part exactly. It held every well at 95 °C and 55 °C in turn, read the light in each cycle and reported the cycle at which the light crossed the threshold, for the false product as faithfully as for the virus.
Neither the chemistry nor the instrument could decide when a count should be believed. The regulatory system decided, outside the tube. The instructions the FDA authorized set the controls and the cut-off, each laboratory had to verify the kit before reporting a result, and failures went from the laboratories to the CDC and on to the FDA, which reissued the authorization without N3 on 15 March 2020. All of it rested on the declaration of 4 February 2020. The FDA's list still showed the panel in August 2026, and its authorization ceases to be in effect with all the others on that list on 26 December 2026, when the declaration ends.
- Under the IVDR, PCR instruments are class A and self-declared, and non-binding guidance of March 2025 moved professional SARS-CoV-2 tests from class D to class B.
- EU legacy devices may stay on the market until the end of 2027, 2028 or 2029 by class if their makers applied in time, and the 2025 proposal to simplify the rules had not been adopted as of October 2026.
- In India, PCR analyzers are class A and COVID-19 molecular kits class C, every class of diagnostic device needs a license, and new class B to D devices need a performance evaluation on three batches.
- One evidence core serves all three jurisdictions, and in the CDC's panel the chemistry copied, the instrument read and the regulatory system decided when to trust the count, under a declaration that ends on 26 December 2026.
Lessons.
The chapters reduce to a short set of working rules for anyone who designs, verifies, buys, regulates or writes about PCR instruments and tests. Each one traces back to a mechanism explained earlier.
- Specify sensitivity in copies per reaction. The chemistry, the optics and the statistics of sampling act on the molecules that reach the well, not on the concentration in the specimen (Chapters 1, 6 and 16).
- Add sample, not cycles, to detect rarer targets. Below about three copies per reaction more than one reaction in twenty receives none, and no number of cycles or brightness of dye changes that (Chapters 6 and 21).
- Check every oligonucleotide against every other, at the 3′ ends above all. Any two that can pair and be extended can make a product in any tube, including those with no target (Chapters 5 and 22).
- Read quantity in the exponential phase or count it in partitions, never from one reaction's plateau. The plateau forgets the starting amount; the cycle at which a curve rises remembers it, and so does the fraction of partitions that stay dark (Chapters 8, 11, 13 and 14).
- State efficiency with every quantity. A tenfold difference is 3.32 cycles only at perfect doubling, and unequal efficiencies masquerade as unequal amounts (Chapters 6 and 13).
- Never compare Ct values across tests, instruments or thresholds. A Ct is a property of one validated test system, not a unit of viral load (Chapter 13).
- Measure temperature in the wells, and time holds from the sample's arrival. The controller reads the block; specifications need their definitions and conditions, and real instruments can start a hold while the liquid is still short of the set point (Chapters 9 and 10).
- Trace every reported number back through its conversions. Copies become light, electrons, counts, corrected values, a curve, a Cq and a call, and each step needs its own calibration and check (Chapters 12 and 15).
- Own the interfaces. Plate, seal, volume, dyes, read timing, data formats and software version are where instruments and assays fail together (Chapters 15 and 17).
- Freeze the test system and treat any change as a change to the test. Instrument, software, reagents, extraction, settings and rules were validated together and must be re-shown together (Chapters 17 and 22).
- Put an internal control in every specimen. It turns a poor swab, a failed extraction or an inhibitor from a false negative into an invalid result, and a cold well too when it shares the target's well (Chapters 19, 20 and 21).
- Read predictive value with the prevalence of the people tested. The same test can be right about two positives in three in one population and about forty-nine in fifty in another (Chapter 21).
- Build one evidence core and present it to each regulator. Analytical and clinical studies, risk, software and safety evidence serve the US, the EU and India; keep filed, cleared, granted, authorized, approved and certified apart (Chapters 19, 25 and 26).
- Read a product's status in the regulator's record, with its date. Cleared, granted, waived and authorized for emergency use are different records, a company's statement is not one of them, and an emergency authorization lasts only as long as the declaration behind it (Chapters 24 and 25).
Glossary.
Terms are defined as they are used in this guide.
- 510(k)
- A US premarket notification showing that a device is substantially equivalent to a legally marketed device; a cleared 510(k) allows marketing.
- Accuracy (temperature)
- How close the average temperature of the wells is to the set point; distinct from uniformity.
- Amplicon
- The stretch of DNA between and including the two primers, which PCR copies exponentially; also the product itself.
- Analyte
- The substance or marker a test detects or measures, such as a virus's RNA in a respiratory specimen.
- Analytical performance
- A test's ability to detect or measure its analyte correctly: limit of detection, specificity, precision and related characteristics.
- Annealing
- The re-pairing of complementary strands on cooling; in PCR, the step in which primers bind their targets.
- Baseline
- The background fluorescence of the early cycles, subtracted from each curve before a threshold is applied.
- BCR-ABL1
- The fusion gene of chronic myeloid leukemia, formed by an exchange between chromosomes 9 and 22, whose RNA is measured to monitor treatment.
- cDNA
- Complementary DNA: DNA copied from an RNA template by reverse transcriptase.
- CDSCO
- The Central Drugs Standard Control Organisation, which publishes the class of each diagnostic device type in India and through its central licensing authority licenses the manufacture of class C and D devices.
- CLIA
- The US Clinical Laboratory Improvement Amendments, which regulate clinical laboratories and sort tests into waived, moderate-complexity and high-complexity categories.
- Clinical performance
- How well a test's results correspond to the clinical condition in the intended population: sensitivity, specificity and predictive values.
- Clinical utility
- Whether using a test improves outcomes for patients; it matters to doctors and payers but is generally not required for market authorization.
- Common specifications
- EU requirements set by implementing regulation for the performance of certain class D diagnostic devices.
- Cq (Ct)
- The fractional cycle at which a well's fluorescence curve crosses the threshold; MIQE's quantification cycle, called threshold cycle (Ct) or crossing point (Cp) by some manufacturers.
- CRISPR detection
- Detection using a guided bacterial enzyme that, on binding its target sequence, cuts nearby reporter molecules to produce a signal.
- De Novo
- The US route for a new type of low-to-moderate-risk device with no predicate; a granted request creates a new classification.
- Denaturation
- Separating the two strands of DNA, in PCR by heating to 94 to 98 °C.
- Digital PCR
- PCR in which the sample is divided into thousands of partitions that are each read as positive or negative, giving an absolute count of molecules.
- dNTP
- Deoxynucleoside triphosphate: one of the four building blocks the polymerase adds to a growing DNA strand.
- Dual Track
- In the FDA's 510(k) records, a submission that combines a 510(k) with an application for a CLIA waiver.
- dUTP and UNG
- A contamination defense: products are made with uracil, and uracil-DNA glycosylase destroys any carried-over product before the next PCR.
- Efficiency
- The fraction of target molecules copied in one cycle; at 100 % the target doubles each cycle.
- Emergency Use Authorization (EUA)
- US authorization of an unapproved product during a declared emergency, lasting until the declaration ends or the authorization is revoked.
- Endpoint PCR
- PCR in which the product is examined only after the last cycle, typically on a gel.
- Error budget
- The allocation of a requirement, such as a temperature tolerance, among the separate sources of error that contribute to it.
- EU reference laboratory
- A laboratory designated under the IVD Regulation to verify the performance of class D devices and their compliance with common specifications, and to test their batches.
- Exclusivity
- A test's failure to react with organisms other than its target; tested as cross-reactivity.
- Extension
- The step in which the polymerase lengthens a bound primer along its template.
- Extraction
- Releasing and purifying nucleic acid from a specimen, removing inhibitors and often concentrating it.
- Fluorophore
- A molecule that absorbs light of one color and emits light of a longer wavelength.
- Heated lid
- The plate above the tubes held at about 105 °C to prevent condensation and to press the tubes into the block.
- Hot start
- Any method that keeps the polymerase inactive until the first heating step, to prevent wrong products forming during set-up.
- Hydrolysis probe
- A probe carrying a fluorophore and a quencher that the polymerase cuts as it copies the target, releasing fluorescence; often called a TaqMan probe.
- Inclusivity
- A test's ability to detect all variants of its target organism.
- Inhibitor
- A substance carried in with a specimen or from extraction that slows or stops amplification.
- Instrument family
- Instruments from one manufacturer with the same general architecture, design, tolerances and capabilities, for the purpose of extending a cleared assay.
- Intended use
- The written statement of what a device measures, in what specimen, for whom, by whom, where and for what decision.
- Interface
- A boundary between two parts of a system, with parameters both sides must agree on.
- Internal control
- A target amplified from every specimen to show that the specimen and extraction were adequate, and amplification too in any well it shares; in the CDC test, RNase P, in a well of its own.
- Isothermal amplification
- Copying nucleic acid at a single temperature, using enzymes that separate strands as they copy.
- IVDR
- Regulation (EU) 2017/746 on in vitro diagnostic medical devices, applicable since 26 May 2022.
- Laboratory-developed test
- A test made and used within one laboratory.
- LAMP
- Loop-mediated isothermal amplification: amplification at about 65 °C with a strand-displacing polymerase and four to six primers.
- Legacy device
- In the EU, a diagnostic device placed on the market under the directive that preceded the IVD Regulation, allowed to stay on the market until a deadline set by its class if its maker applied to a notified body in time.
- Limit of blank
- The highest result expected from a sample containing no analyte.
- Limit of detection
- For a qualitative molecular test, the lowest concentration detected in at least 95 % of replicates.
- Long product
- A strand made in the early cycles from an original template, with one end defined by a primer and the other ragged.
- Lower limit of quantification
- The smallest amount a quantitative test can measure with acceptable precision.
- Master mix
- The ready-made mixture of polymerase, dNTPs, magnesium and buffer to which primers, probes and sample are added.
- MDCG 2020-16
- The Medical Device Coordination Group's guidance on applying the IVD Regulation's classification rules; it is not legally binding.
- Melt curve
- A plot of how fluorescence falls as the product is heated slowly; peaks mark the melting temperatures of the products present.
- Melting temperature (Tm)
- The temperature at which half of a DNA duplex has separated into single strands; it depends on sequence, salt, magnesium and strand concentration.
- MIQE
- Minimum Information for Publication of Quantitative Real-Time PCR Experiments: reporting guidelines published in 2009 and revised in 2025.
- Mispriming
- A primer binding to and being extended from a partly matching site.
- Multicomponent analysis
- Separating the contributions of several dyes to measured light using each dye's calibrated spectrum.
- Multiplex
- Detecting several targets in one reaction, usually with a different fluorophore for each.
- No-template control
- A reaction with every reagent but no sample, which should show no amplification; any signal reveals contamination or a product the reagents make on their own, such as a primer-dimer.
- Notified body
- An organization designated in the EU to assess the conformity of devices in classes that require third-party assessment.
- Oligonucleotide (oligo)
- A short synthetic strand of DNA; primers and probes are oligonucleotides.
- Open platform
- A real-time PCR instrument that runs any assay written for it, a kit maker's or a laboratory's own.
- Partition
- One of the thousands of small separate reactions into which digital PCR divides a sample.
- Passive reference
- An inert dye such as ROX added to every well, by which the reporter signal is divided to correct for volume and optical differences.
- Peltier module
- A thermoelectric device that pumps heat from one face to the other in the direction set by its current.
- Performance evaluation (India)
- Testing of a new class B, C or D diagnostic device on three batches made from different lots of key raw materials, by a laboratory set up or designated by the central government or accredited by the national accreditation board.
- Plateau
- The final phase of a PCR, in which product stops increasing; reached at similar levels from very different starting amounts.
- Poisson distribution
- The statistics of items scattered at random, which govern how many molecules land in a reaction or partition.
- Polymerase
- The enzyme that builds a new DNA strand by adding nucleotides to a primer, copying a template.
- Positive and negative percent agreement
- Agreement of a new test with a comparator that is not a reference standard; not to be reported as sensitivity and specificity.
- Positive control
- A reaction containing the target at a known level, which must be positive for the run to be valid.
- Predicate
- The legally marketed device to which a 510(k) shows a new device to be substantially equivalent.
- Predictive value
- The probability that a positive (or negative) result is correct; it depends on the prevalence in the people tested.
- Prequalification (WHO)
- WHO's assessment of a named diagnostic product, separate from a WHO recommendation that a type of test may be used for a purpose.
- Primer
- A short designed piece of DNA that binds the template and gives the polymerase a starting point; PCR uses two, facing each other.
- Primer-dimer
- A short product formed when two primers pair with each other and are extended.
- Probe
- A short labeled oligonucleotide that binds inside the target and reports its presence by fluorescence.
- QMSR
- The US Quality Management System Regulation, 21 CFR 820, in force since 2 February 2026, incorporating ISO 13485:2016.
- Quencher
- A molecule that absorbs a nearby fluorophore's energy and keeps it dark until they are separated.
- Ramp rate
- The rate of temperature change, in degrees per second; defined variously for the block or the sample, as a maximum or an average.
- Real-time PCR
- PCR in which fluorescence is read in every cycle, so the cycle at which the signal rises can be used to infer the starting amount.
- Research use only
- A label for products that may not be used to diagnose patients.
- Reverse transcriptase
- An enzyme that copies RNA into complementary DNA.
- RNase
- An enzyme that degrades RNA; robust and widespread.
- RNase P
- A human gene amplified as the CDC test's internal control.
- RPA
- Recombinase polymerase amplification: isothermal amplification at about 37 °C using recombination proteins to insert primers.
- Sample-temperature model
- The controller's calculation of the liquid's temperature from the block's history, the volume and the tube's time constant.
- Scientific validity
- The association of an analyte with a clinical condition; in the EU's IVD Regulation, the first of the three questions a diagnostic's evidence answers.
- Sensitivity (clinical)
- The fraction of people with the condition whom a test calls positive, measured against a reference standard.
- Special controls
- Requirements written into a US classification regulation that every class II device of that type must meet, such as the studies and labeling required for SARS-CoV-2 multi-target tests.
- Specificity (clinical)
- The fraction of people without the condition whom a test calls negative, measured against a reference standard.
- Stacking
- The attraction between neighboring base pairs along the helix, which provides most of a duplex's stability.
- Standard curve
- A plot of Cq against the logarithm of known starting amounts, used to convert an unknown's Cq into an amount and to estimate efficiency.
- Standard evaluation protocol
- A protocol published by ICMR and CDSCO for evaluating one type of diagnostic kit; a compendium of 39 was released in September 2025.
- Syndromic panel
- A test that checks one specimen for the many pathogens that can cause the same symptoms, such as a respiratory infection.
- Taq polymerase
- The heat-stable DNA polymerase of the bacterium Thermus aquaticus, used in most PCR since 1988.
- Test system
- The assay, instrument and software together, evaluated and authorized as one.
- Thermal cycler
- The instrument that heats and cools PCR reactions on a programmed schedule; often called a PCR machine.
- Threshold
- The fluorescence level, set above background in the exponential region, at which Cq is read.
- Time constant
- For a first-order lag, the time the sample takes to cover about 63 % of a step in temperature.
- Uniformity (temperature)
- How closely the wells of a block agree with one another at a set point; distinct from accuracy.
- V model
- The systems-engineering picture of decomposing needs into requirements and designs, then verifying and validating upward.
- Validation
- Showing that a device meets its users' needs and intended use.
- Verification
- Showing that a design's outputs meet its inputs: that the product was built as specified.
- Waived test
- Under CLIA, a test that meets separate criteria of simplicity and low risk of error; other tests are scored as moderate or high complexity.
Sources.
Sources are listed by the chapter in which they are first used. Figures from manufacturers, and from studies written by manufacturers' staff, are identified as such in the text. Regulatory status and standards editions are as of October 2026.
- Opening — US Food and Drug Administration. Letter to Robert R. Redfield, Director, CDC, reissuing the Emergency Use Authorization for the CDC 2019-nCoV Real-Time RT-PCR Diagnostic Panel (1 Dec 2020), with the revision history from 4 Feb 2020. https://www.fda.gov/media/134919/download
- Opening — CDC. Shipping of CDC 2019 novel coronavirus diagnostic test kits begins (press release, 6 Feb 2020). archive.cdc.gov/www_cdc_gov/media/releases/2020/p0206-coronavirus-diagnostic-test-kits.html
- Opening — CDC. Transcripts of the CDC telebriefings on the novel coronavirus, 5, 12 and 28 Feb 2020.
- Opening — NPR. CDC report: officials knew coronavirus test was flawed but released it anyway (6 Nov 2020).
- Opening — Lee J.S., Goldstein J.M., Moon J.L., et al. Analysis of the initial lot of the CDC 2019-Novel Coronavirus (2019-nCoV) real-time RT-PCR diagnostic panel. PLoS One 16, e0260487 (2021). doi:10.1371/journal.pone.0260487
- Opening — 42 CFR 493.1253, Standard: Establishment and verification of performance specifications (eCFR).
- Ch. 1 — Nurk S., Koren S., Rhie A., et al. The complete sequence of a human genome. Science 376, 44–53 (2022). doi:10.1126/science.abj6987
- Ch. 1 — New England Biolabs. Nucleic acid data (web page). neb.com/en/tools-and-resources/usage-guidelines/nucleic-acid-data
- Ch. 1 — Thermo Fisher Scientific. Creating standard curves with genomic DNA or plasmid DNA templates for use in quantitative PCR (application note). Manufacturer.
- Ch. 1 — Thermo Fisher Scientific. Ethidium bromide: detection limit (product page). Manufacturer.
- Ch. 1 — Bustin S.A., Ruijter J.M., van den Hoff M.J.B., et al. MIQE 2.0: revision of the Minimum Information for Publication of Quantitative Real-Time PCR Experiments guidelines. Clin Chem 71, 634–651 (2025). doi:10.1093/clinchem/hvaf043
- Ch. 1 — CDC. CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel: Instructions for Use. CDC-006-00019, Rev. 08, effective 7 Mar 2023. fda.gov/media/134922/download
- Ch. 1 — Lu X., Wang L., Sakthivel S.K., et al. US CDC real-time reverse transcription PCR panel for detection of severe acute respiratory syndrome coronavirus 2. Emerg Infect Dis 26(8) (2020). doi:10.3201/eid2608.201246
- Ch. 2 — Yakovchuk P., Protozanova E., Frank-Kamenetskii M.D. Base-stacking and base-pairing contributions into thermal stability of the DNA double helix. Nucleic Acids Res 34, 564–574 (2006). doi:10.1093/nar/gkj454
- Ch. 2 — Marmur J., Doty P. Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature. J Mol Biol 5, 109–118 (1962). doi:10.1016/S0022-2836(62)80066-7
- Ch. 2 — Integrated DNA Technologies. Understanding melting temperature (Tm) (web article). Manufacturer.
- Ch. 2 — Potapov V., Ong J.L. Examining sources of error in PCR by single-molecule sequencing. PLoS One 12, e0169774 (2017). doi:10.1371/journal.pone.0169774 (authors employed by New England Biolabs)
- Ch. 3 — New England Biolabs. Guidelines for PCR optimization with Taq DNA polymerase (web page). Manufacturer.
- Ch. 3 — Promega. GoTaq PCR Core Systems, Technical Bulletin TB254 (rev. 5/19). Manufacturer.
- Ch. 3 — Saiki R.K., Scharf S., Faloona F., et al. Enzymatic amplification of β-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230, 1350–1354 (1985). doi:10.1126/science.2999980
- Ch. 3 — Arizona State University Embryo Project Encyclopedia. Polymerase chain reaction (PCR). embryo.asu.edu
- Ch. 3 — Spahich N. A not-so-simple idea. The Scientist (1 June 2023).
- Ch. 3 — Nobel Prize Outreach. The Nobel Prize in Chemistry 1993 (summary) and Mullis K.B., The polymerase chain reaction, Nobel lecture (8 Dec 1993). nobelprize.org
- Ch. 3 — Mullis K.B. The unusual origin of the polymerase chain reaction. Sci Am 262(4), 56–65 (1990). doi:10.1038/scientificamerican0490-56 (read in a reproduction)
- Ch. 4 — Innis M.A., Myambo K.B., Gelfand D.H., Brow M.A.D. DNA sequencing with Thermus aquaticus DNA polymerase and direct sequencing of polymerase chain reaction-amplified DNA. PNAS 85, 9436–9440 (1988). doi:10.1073/pnas.85.24.9436 (authors at Cetus; rates as given in the companion US Patent 5,075,216)
- Ch. 4 — Lawyer F.C., Stoffel S., Saiki R.K., et al. High-level expression, purification, and enzymatic characterization of full-length Thermus aquaticus DNA polymerase and a truncated form deficient in 5′ to 3′ exonuclease activity. PCR Methods Appl 2, 275–287 (1993). doi:10.1101/gr.2.4.275 (authors at Roche Molecular Systems)
- Ch. 4 — Eckert K.A., Kunkel T.A. High fidelity DNA synthesis by the Thermus aquaticus DNA polymerase. Nucleic Acids Res 18, 3739–3744 (1990). doi:10.1093/nar/18.13.3739
- Ch. 4 — Medtech Insight. Roche/Perkin-Elmer Taq polymerase licensing deal with Boehringer Mannheim (26 Jul 1993). insights.citeline.com
- Ch. 4 — The Pharma Letter. Roche loosens PCR stranglehold (8 Aug 1993). thepharmaletter.com
- Ch. 5 — Chou Q., Russell M., Birch D.E., Raymond J., Bloch W. Prevention of pre-PCR mis-priming and primer dimerization improves low-copy-number amplifications. Nucleic Acids Res 20, 1717–1723 (1992). doi:10.1093/nar/20.7.1717 (authors at Cetus)
- Ch. 5 — Kellogg D.E., Rybalkin I., Chen S., et al. TaqStart antibody: hot start PCR facilitated by a neutralizing monoclonal antibody directed against Taq DNA polymerase. BioTechniques 16, 1134–1137 (1994).
- Ch. 5 — Birch D.E., Kolmodin L., Wong J., et al. Simplified hot start PCR. Nature 381, 445 (1996). doi:10.1038/381445a0
- Ch. 5 — Dang C., Jayasena S.D. Oligonucleotide inhibitors of Taq DNA polymerase facilitate detection of low copy number targets by PCR. J Mol Biol 264, 268–278 (1996). doi:10.1006/jmbi.1996.0640
- Ch. 5 — Carr S.M. The polymerase chain reaction (in detail). Memorial University of Newfoundland (teaching page). mun.ca/biology/scarr/PCR_in_detail.html
- Ch. 5 — Thermo Fisher Scientific. PCR primer design tips. Behind the Bench (25 Sep 2019). Manufacturer.
- Ch. 5 — FDA. Molecular diagnostic template for SARS-CoV-2 emergency use authorization requests (version of 22 Sep 2020), as reproduced by PRoDTEC. fda.gov/media/135900/download
- Ch. 5 — Batzer M.A., Deininger P.L. Alu repeats and human genomic diversity. Nat Rev Genet 3, 370–379 (2002). doi:10.1038/nrg798
- Ch. 6 — Bustin S.A., Benes V., Garson J.A., et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55, 611–622 (2009). doi:10.1373/clinchem.2008.112797
- Ch. 7 — Brock T.D., Freeze H. Thermus aquaticus gen. n. and sp. n., a nonsporulating extreme thermophile. J Bacteriol 98, 289–297 (1969). doi:10.1128/jb.98.1.289-297.1969
- Ch. 7 — US Geological Survey, Yellowstone Volcano Observatory. How a thermophilic bacterium from a Yellowstone hot spring is helping fight against COVID-19 (web article). usgs.gov
- Ch. 7 — Chien A., Edgar D.B., Trela J.M. Deoxyribonucleic acid polymerase from the extreme thermophile Thermus aquaticus. J Bacteriol 127, 1550–1557 (1976). doi:10.1128/jb.127.3.1550-1557.1976
- Ch. 7 — Saiki R.K., Gelfand D.H., Stoffel S., et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487–491 (1988). doi:10.1126/science.2448875 (authors at Cetus)
- Ch. 7 — Smithsonian Institution Archives. History of the Polymerase Chain Reaction Videohistory Collection (RU 9577). si.edu
- Ch. 7 — Thermo Fisher Scientific. 30 years of thermal cycler innovations #1: the world's first thermal cycler. Behind the Bench (blog). Manufacturer.
- Ch. 7 — Cline J., Braman J.C., Hogrefe H.H. PCR fidelity of Pfu DNA polymerase and other thermostable DNA polymerases. Nucleic Acids Res 24, 3546–3551 (1996).
- Ch. 7 — Holland P.M., Abramson R.D., Watson R., Gelfand D.H. Detection of specific polymerase chain reaction product by utilizing the 5′→3′ exonuclease activity of Thermus aquaticus DNA polymerase. PNAS 88, 7276–7280 (1991). doi:10.1073/pnas.88.16.7276 (authors at Cetus)
- Ch. 7 — Sigma-Aldrich (Merck). Antibody/enzyme-mediated hot start PCR (technical page). Manufacturer.
- Ch. 8 — Morrison C., Gannon F. The impact of the PCR plateau phase on quantitative PCR. Biochim Biophys Acta 1219, 493–498 (1994). doi:10.1016/0167-4781(94)90076-0
- Ch. 8 — Kainz P. The PCR plateau phase – towards an understanding of its limitations. Biochim Biophys Acta 1494, 23–27 (2000). doi:10.1016/S0167-4781(00)00200-1
- Ch. 8 — US National Institute of Justice. DNA amplification online training: plateau effect. nij.ojp.gov
- Ch. 9 — Ferrotec. Thermoelectric technical reference: mathematical modeling of thermoelectric cooling modules (section 11) and reliability of thermoelectric cooling modules (section 10). thermal.ferrotec.com. Manufacturer.
- Ch. 9 — Barako M.T., Park W., Marconnet A.M., Asheghi M., Goodson K.E. Thermal cycling, mechanical degradation, and the effective figure of merit of a thermoelectric module. J Electron Mater 42, 372–381 (2013). doi:10.1007/s11664-012-2366-1
- Ch. 9 — Roche Diagnostics. LightCycler 480 Instrument Operator's Manual, software 1.5, rev. 3.0 (2008). Manufacturer.
- Ch. 9 — Atwood J.G., Mossa A.C., Goven L.M., et al. Thermal cycler for automatic performance of the polymerase chain reaction with close temperature control. US Patent 5,475,610 (filed 20 Apr 1992; issued 12 Dec 1995).
- Ch. 9 — National Institute of Advanced Industrial Science and Technology (AIST). Thermophysical property database: thermal conductivity of metals. tpds.db.aist.go.jp
- Ch. 9 — Thermo Fisher Scientific. What is the max block ramp rate of the GeneAmp PCR System 9700? (product FAQ). Manufacturer.
- Ch. 9 — Bio-Rad. CFX96 Touch Real-Time PCR Detection System, Bulletin 6093 Ver F (2020). Manufacturer.
- Ch. 9 — Wittwer C.T., Fillmore G.C., Hillyard D.R. Automated polymerase chain reaction in capillary tubes with hot air. Nucleic Acids Res 17, 4353–4357 (1989). doi:10.1093/nar/17.11.4353
- Ch. 9 — Roche Diagnostics. LightCycler 2.0 Instrument (product page). Manufacturer.
- Ch. 9 — QIAGEN. Rotor-Gene Q MDx (US) (product page). Manufacturer.
- Ch. 9 — Kopp M.U., de Mello A.J., Manz A. Chemical amplification: continuous-flow PCR on a chip. Science 280, 1046–1048 (1998). doi:10.1126/science.280.5366.1046
- Ch. 9 — Krishnan M., Ugaz V.M., Burns M.A. PCR in a Rayleigh–Bénard convection cell. Science 298, 793 (2002).
- Ch. 9 — Son J.H., Cho B., Hong S., et al. Ultrafast photonic PCR. Light Sci Appl 4, e280 (2015). doi:10.1038/lsa.2015.53
- Ch. 9 — Ahrberg C.D., Manz A., Chung B.G. Polymerase chain reaction in microfluidic devices. Lab Chip 16, 3866–3884 (2016). doi:10.1039/C6LC00984K
- Ch. 10 — Thermo Fisher Scientific. Applied Biosystems 7500 Fast Dx Real-Time PCR Instrument (brochure CO016365, 2015). Manufacturer.
- Ch. 10 — Bio-Rad. C1000 Touch Thermal Cycler, Bulletin 6095 Ver C (2019). Manufacturer.
- Ch. 10 — Bio-Rad. CFX Opus 96 Real-Time PCR System, Bulletin 7299 Ver D (2023). Manufacturer.
- Ch. 10 — Thermo Fisher Scientific. QuantStudio 5 Real-Time PCR System, specification sheet CO014799 (2015). Manufacturer.
- Ch. 10 — Eppendorf. Mine is the fastest cycler ever – really? Lab Academy (18 Sep 2018). Manufacturer.
- Ch. 10 — Schoder D., Schmalwieser A., Schauberger G., et al. Novel approach for assessing performance of PCR cyclers used for diagnostic testing. J Clin Microbiol 43, 2724–2728 (2005). doi:10.1128/JCM.43.6.2724-2728.2005
- Ch. 10 — Thermo Fisher Scientific. Thermal cycler temperature accuracy: a comparison of several models (application note APN-11799200). Manufacturer.
- Ch. 10 — METTLER TOLEDO. PCR thermal cycler calibration (service page). mt.com
- Ch. 10 — Span M., Verblakt M., Hendrikx T. Measurement uncertainty in calibration and compliancy testing of PCR and qPCR thermal cyclers. 18th International Congress of Metrology, 05003 (2017). doi:10.1051/metrology/201705003 (first and third authors at CYCLERtest BV; second author at CelsiusLabs, at the same address)
- Ch. 11 — Higuchi R., Dollinger G., Walsh P.S., Griffith R. Simultaneous amplification and detection of specific DNA sequences. Bio/Technology 10, 413–417 (1992). doi:10.1038/nbt0492-413
- Ch. 11 — Higuchi R., Fockler C., Dollinger G., Watson R. Kinetic PCR analysis: real-time monitoring of DNA amplification reactions. Bio/Technology 11, 1026–1030 (1993). doi:10.1038/nbt0993-1026
- Ch. 11 — Heid C.A., Stevens J., Livak K.J., Williams P.M. Real time quantitative PCR. Genome Res 6, 986–994 (1996). doi:10.1101/gr.6.10.986 (authors at Applied Biosystems/Perkin-Elmer and Genentech)
- Ch. 11 — Sykes P.J., Neoh S.H., Brisco M.J., et al. Quantitation of targets for PCR by use of limiting dilution. BioTechniques 13, 444–449 (1992).
- Ch. 11 — Vogelstein B., Kinzler K.W. Digital PCR. PNAS 96, 9236–9241 (1999). doi:10.1073/pnas.96.16.9236
- Ch. 11 — BioWorld. Cetus to sell PCR technology (23 Jul 1991); Roche, Perkin-Elmer form alliance (12 Dec 1991). bioworld.com
- Ch. 12 — Zipper H., Brunner H., Bernhagen J., Vitzthum F. Investigations on DNA intercalation and surface binding by SYBR Green I, its structure determination and methodological implications. Nucleic Acids Res 32, e103 (2004). doi:10.1093/nar/gnh101
- Ch. 12 — Integrated DNA Technologies. Interpreting melt curves: an indicator, not a diagnosis (web article). Manufacturer.
- Ch. 12 — Lee L.G., Connell C.R., Bloch W. Allelic discrimination by nick-translation PCR with fluorogenic probes. Nucleic Acids Res 21, 3761–3766 (1993). doi:10.1093/nar/21.16.3761 (authors at Applied Biosystems/Perkin-Elmer)
- Ch. 12 — Tyagi S., Kramer F.R. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol 14, 303–308 (1996). doi:10.1038/nbt0396-303
- Ch. 12 — Applied Biosystems. 7300/7500/7500 Fast Real-Time PCR System Installation and Maintenance Guide, P/N 4378657 Rev. A (2006). Manufacturer.
- Ch. 12 — Thermo Fisher Scientific. QuantStudio 5 Dx Real-Time PCR System, specification sheet COL25645 (2022). Manufacturer.
- Ch. 12 — Shu B., Kirby M.K., Davis W.G., et al. Multiplex real-time reverse transcription PCR for influenza A virus, influenza B virus, and severe acute respiratory syndrome coronavirus 2. Emerg Infect Dis 27(7) (2021). doi:10.3201/eid2707.210462
- Ch. 12 — Thermo Fisher Scientific. ROX passive reference dye for troubleshooting real-time PCR (application note CO016741, 2015). Manufacturer.
- Ch. 12 — CDC. Real-time RT-PCR panel for detection 2019-novel coronavirus: instructions for use (effective 24 Jan 2020). stacks.cdc.gov/view/cdc/84526
- Ch. 12 — Thermo Fisher Scientific. TaqPath 1-Step RT-qPCR Master Mix, CG, cat. A15299 and A15300 (product page). Manufacturer.
- Ch. 13 — QIAGEN. Commonly used terms in PCR (web page). Manufacturer.
- Ch. 13 — Verification of real-time PCR methods for qualitative and quantitative testing of genetically modified organisms (2012), quoting the minimum performance requirements of the European Network of GMO Laboratories. Secondary source.
- Ch. 13 — Livak K.J., Schmittgen T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2^−ΔΔCT method. Methods 25, 402–408 (2001). doi:10.1006/meth.2001.1262 (first author at Applied Biosystems)
- Ch. 13 — Public Health Agency of Canada. COVID-19 testing: polymerase chain reaction (PCR) and cycle threshold (Ct) values (modified 11 Sep 2026). canada.ca
- Ch. 13 — AACC. AACC recommendation for reporting SARS-CoV-2 cycle threshold (Ct) values (17 Mar 2021). aacc.org
- Ch. 13 — Association for Molecular Pathology and Infectious Diseases Society of America. Important issues to consider before interpreting and applying Ct values in clinical practice (2021). amp.org
- Ch. 13 — Bullard J., Dust K., Funk D., et al. Predicting infectious severe acute respiratory syndrome coronavirus 2 from diagnostic samples. Clin Infect Dis 71, 2663–2666 (2020). doi:10.1093/cid/ciaa638
- Ch. 13 — Singanayagam A., Patel M., Charlett A., et al. Duration of infectiousness and correlation with RT-PCR cycle threshold values in cases of COVID-19, England, January to May 2020. Euro Surveill 25, 2001483 (2020). doi:10.2807/1560-7917.ES.2020.25.32.2001483
- Ch. 14 — Majumdar N., Wessel T., Marks J. Digital PCR modeling for maximal sensitivity, dynamic range and measurement precision. PLoS One 10, e0118833 (2015). doi:10.1371/journal.pone.0118833 (authors employed by Thermo Fisher Scientific)
- Ch. 14 — Corbisier P., Pinheiro L., Mazoua S., et al. DNA copy number concentration measured by digital and droplet digital quantitative PCR using certified reference materials. Anal Bioanal Chem 407, 1831–1840 (2015). doi:10.1007/s00216-015-8458-z
- Ch. 14 — Hindson B.J., Ness K.D., Masquelier D.A., et al. High-throughput droplet digital PCR system for absolute quantitation of DNA copy number. Anal Chem 83, 8604–8610 (2011). doi:10.1021/ac202028g (authors mostly at Bio-Rad, which had acquired QuantaLife)
- Ch. 14 — Bio-Rad. QX200 Droplet Digital PCR System (product page) and QX600 Droplet Digital PCR System, Bulletin 3557 Ver A (2023). Manufacturer.
- Ch. 14 — QIAGEN. QIAcuity Digital PCR System and QIAcuity Nanoplates (product pages). Manufacturer.
- Ch. 14 — Thermo Fisher Scientific. QuantStudio Absolute Q Digital PCR System, specification sheet COL25472 (2021). Manufacturer.
- Ch. 14 — Roche. Roche Digital LightCycler System: a technological guide (2022). Manufacturer.
- Ch. 14 — Bio-IT World. Bio-Rad expands Droplet Digital PCR offering through strategic acquisition and platform rollout (9 Jul 2025).
- Ch. 14 — Sidstedt M., Rådström P., Hedman J. PCR inhibition in qPCR, dPCR and MPS — mechanisms and solutions. Anal Bioanal Chem (2020). doi:10.1007/s00216-020-02490-2
- Ch. 14 — Bio-Rad Laboratories. Annual report on Form 10-K for 2011: acquisition of QuantaLife, Inc. (October 2011). sec.gov
- Ch. 14 — Bio-Rad. QX700 Droplet Digital PCR System Instrument Guide, 10000171493 (Nov 2025). bio-rad.com/sites/default/files/2025-11/10000171493.pdf (manufacturer)
- Ch. 15 — FDA. Content of Premarket Submissions for Device Software Functions (final guidance, 14 Jun 2023). fda.gov/media/153781/download
- Ch. 16 — Mooz H., Forsberg K. The relationship of system engineering to the project cycle. Joint conference of NCOSE and the American Society for Engineering Management (1991); INCOSE International Symposium 1(1). doi:10.1002/j.2334-5837.1991.tb01484.x
- Ch. 16 — FDA. Design Control Guidance for Medical Device Manufacturers (11 Mar 1997).
- Ch. 16 — FDA. Medical Devices; Quality System Regulation Amendments. Final rule, 89 FR 7496 (2 Feb 2024), effective 2 Feb 2026; and FDA, Quality Management System Regulation (QMSR) (web page).
- Ch. 16 — ISO 13485:2016. Medical devices — Quality management systems — Requirements for regulatory purposes (confirmed 2025).
- Ch. 17 — FDA. Assay Migration Studies for In Vitro Diagnostic Devices (final guidance, April 2013). fda.gov/media/73669/download
- Ch. 17 — FDA. Replacement Reagent and Instrument Family Policy for In Vitro Diagnostic Devices (final guidance, 17 Aug 2022). fda.gov/media/111186/download
- Ch. 18 — Raja S., Ching J., Xi L., et al. Technology for automated, rapid, and quantitative PCR or reverse transcription-PCR clinical testing. Clin Chem 51, 882–890 (2005). doi:10.1373/clinchem.2004.046474 (four authors at Cepheid)
- Ch. 18 — Cepheid. Technology of the GeneXpert System, training document 301-7932 Rev B.1 (2020). Manufacturer.
- Ch. 18 — FDA. Emergency Use Authorization letter for Xpert Xpress SARS-CoV-2 (first issued 20 Mar 2020); Cepheid, Xpert Xpress SARS-CoV-2 Instructions for Use, Rev. G (2022); Cepheid press release (21 Mar 2020). Manufacturer for time to result.
- Ch. 18 — Poritz M.A., Blaschke A.J., Byington C.L., et al. FilmArray, an automated nested multiplex PCR system for multi-pathogen detection: development and application to respiratory tract infection. PLoS One 6, e26047 (2011). doi:10.1371/journal.pone.0026047 (authors at Idaho Technology; part-funded by the company)
- Ch. 18 — FDA. 510(k) decision summary K243406: cobas liat SARS-CoV-2, Influenza A/B & RSV (2025).
- Ch. 18 — bioMérieux. BioFire Respiratory 2.1 (RP2.1) Panel (product page). Manufacturer.
- Ch. 18 — Chin C.D., Linder V., Sia S.K. Commercialization of microfluidic point-of-care diagnostic devices. Lab Chip 12, 2118–2134 (2012). doi:10.1039/C2LC21204H
- Ch. 18 — Abbott. ID NOW COVID-19 Emergency Use Authorization Instructions for Use, Rev. 8 (Aug 2021). fda.gov/media/136525/download
- Ch. 19 — IEC 62304:2006+AMD1:2015. Medical device software — Software life cycle processes; and Johner Institute, IEC 62304 second edition (8 Jul 2026).
- Ch. 19 — ISO 14971:2019. Medical devices — Application of risk management to medical devices (confirmed 2025); ISO/TR 24971:2020, Guidance on the application of ISO 14971, Annex H.
- Ch. 19 — 21 CFR 866.3981. Device to detect and identify nucleic acid targets in respiratory specimens from microbial agents that cause the SARS-CoV-2 respiratory infection and other microbial agents (eCFR).
- Ch. 19 — IEC 61010-1:2010+AMD1:2016 CSV; IEC 61010-2-010:2019; IEC 61010-2-081:2019; IEC 61010-2-101:2018 (IEC Webstore); and SIQ, IEC 61010-1 Amendment 2: upcoming changes (3 Apr 2025).
- Ch. 19 — IEC 61326-1:2020 and IEC 61326-2-6:2025. Electrical equipment for measurement, control and laboratory use — EMC requirements (IEC Webstore).
- Ch. 19 — IEC 62366-1:2015+AMD1:2020. Medical devices — Application of usability engineering to medical devices; ISO 18113-3:2022, IVD instruments for professional use (labeling).
- Ch. 19 — FDA. Quality Management System Regulation: frequently asked questions (web page, content current as of 2 Oct 2026). fda.gov/medical-devices/quality-management-system-regulation-qmsr/quality-management-system-regulation-frequently-asked-questions
- Ch. 19 — FDA, CDRH. Risk basics for medical devices (CDRH Learn presentation by J. Tartal, undated), quoting the definitions of ISO 14971:2019 and its Annex C, Figure C.1. fda.gov/media/163912/download (secondary for the standard's wording)
- Ch. 19 — Regulation (EU) 2017/746 on in vitro diagnostic medical devices (IVDR), consolidated version of 10 Jan 2025, EUR-Lex 32017R0746; Annex I, Chapter II, section 16, Electronic programmable systems, read as adopted in the reproduction at legislation.gov.uk/eur/2017/746/annex/I/adopted.
- Ch. 19 — Hoxey E. Updated guidance on implementing risk management for medical devices. BSI Compliance Navigator blog (undated). compliancenavigator.bsigroup.com (secondary, on the diagnostics annex of ISO/TR 24971)
- Ch. 20 — Temin H.M., Mizutani S. RNA-dependent DNA polymerase in virions of Rous sarcoma virus. Nature 226, 1211–1213 (1970). doi:10.1038/2261211a0
- Ch. 20 — Baltimore D. RNA-dependent DNA polymerase in virions of RNA tumour viruses. Nature 226, 1209–1211 (1970). doi:10.1038/2261209a0
- Ch. 20 — Roche Life Science. Reverse transcription technology (web article). Manufacturer.
- Ch. 20 — Thermo Fisher Scientific. What is RT-qPCR? Basic principles (web article). Manufacturer.
- Ch. 20 — Thermo Fisher Scientific (Ambion). The basics: RNase control (technical article). Manufacturer.
- Ch. 20 — Jue E., Witters D., Ismagilov R.F. Two-phase wash to solve the ubiquitous contaminant-carryover problem in commercial nucleic-acid extraction kits. Sci Rep 10, 1940 (2020). doi:10.1038/s41598-020-58586-3
- Ch. 20 — Schrader C., Schielke A., Ellerbroek L., Johne R. PCR inhibitors – occurrence, properties and removal. J Appl Microbiol 113, 1014–1026 (2012). doi:10.1111/j.1365-2672.2012.05384.x
- Ch. 20 — Al-Soud W.A., Rådström P. Purification and characterization of PCR-inhibitory components in blood cells. J Clin Microbiol 39, 485–493 (2001). doi:10.1128/jcm.39.2.485-493.2001
- Ch. 20 — Longo M.C., Berninger M.S., Hartley J.L. Use of uracil DNA glycosylase to control carry-over contamination in polymerase chain reactions. Gene 93, 125–128 (1990). doi:10.1016/0378-1119(90)90145-H (authors at Life Technologies)
- Ch. 20 — Life Technologies, Inc. Annual report on Form 10-K for 1996. sec.gov (company statement)
- Ch. 21 — CLSI. EP17-A2: Evaluation of detection capability for clinical laboratory measurement procedures, 2nd ed. (2012; reaffirmed 2017).
- Ch. 21 — CLSI. EP07: Interference testing in clinical chemistry (current edition).
- Ch. 21 — FDA. Statistical Guidance on Reporting Results from Studies Evaluating Diagnostic Tests (final guidance, 13 Mar 2007).
- Ch. 21 — ISO 15189:2022. Medical laboratories — Requirements for quality and competence; ISO 17822:2020, IVD test systems — Nucleic acid amplification-based examination procedures for detection and identification of microbial pathogens — Laboratory quality practice guide.
- Ch. 21 — CLSI. MM17: Validation and verification of multiplex nucleic acid assays, 2nd ed. (2018); MM19: Establishing molecular testing in medical laboratory environments, 2nd ed. (2025).
- Ch. 22 — Holmes E.C., on behalf of the consortium led by Zhang Y.-Z. Novel 2019 coronavirus genome. virological.org (posted 10/11 Jan 2020).
- Ch. 22 — Corman V.M., Landt O., Kaiser M., et al. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Euro Surveill 25, 2000045 (2020). doi:10.2807/1560-7917.ES.2020.25.3.2000045 (includes an author at TIB Molbiol)
- Ch. 22 — HHS. Determination of public health emergency. 85 FR 7316 (7 Feb 2020).
- Ch. 22 — HHS Office of Inspector General. CDC's internal control weaknesses led to its initial COVID-19 test kit failure, but CDC ultimately created a working test kit. A-04-20-02027 (24 Oct 2023).
- Ch. 22 — Sarata A.K. Early development and regulation of diagnostic testing for COVID-19: frequently asked questions. Congressional Research Service R46261 (updated 17 Mar 2021).
- Ch. 22 — CDC. Lab alerts: Changes to CDC RT-PCR for SARS-CoV-2 testing (21 Jul 2021); Clarifications about retirement of the CDC 2019-nCoV Real-Time RT-PCR Diagnostic Panel (2 Aug 2021).
- Ch. 22 — FDA. In vitro diagnostics EUAs — molecular diagnostic tests for SARS-CoV-2 (web page, content current as of 11 Aug 2026).
- Ch. 22 — HHS. Termination of three declarations authorizing emergency use of medical devices during the COVID-19 pandemic. 91 FR 40544 (2 Jul 2026), effective 26 Dec 2026.
- Ch. 22 — Eurosurveillance editorial team. Editorial note: possible undisclosed conflict of interest. Euro Surveill 25(30), 2007301 (2020). doi:10.2807/1560-7917.ES.2020.25.30.2007301
- Ch. 23 — Farrar J.S., Wittwer C.T. Extreme PCR: efficient and specific DNA amplification in 15–60 seconds. Clin Chem 61, 145–153 (2015). doi:10.1373/clinchem.2014.228304
- Ch. 23 — Makrigiorgos G.M. Extreme PCR meets high-speed melting: a step closer to molecular diagnostics 'while you wait'. Clin Chem 65, 217–219 (2019). doi:10.1373/clinchem.2018.298356
- Ch. 23 — Notomi T., Okayama H., Masubuchi H., et al. Loop-mediated isothermal amplification of DNA. Nucleic Acids Res 28, e63 (2000). doi:10.1093/nar/28.12.e63 (authors at Eiken Chemical)
- Ch. 23 — Piepenburg O., Williams C.H., Stemple D.L., Armes N.A. DNA detection using recombination proteins. PLoS Biol 4, e204 (2006). doi:10.1371/journal.pbio.0040204 (authors mostly at ASM Scientific)
- Ch. 23 — Gootenberg J.S., Abudayyeh O.O., Lee J.W., et al. Nucleic acid detection with CRISPR-Cas13a/C2c2. Science 356, 438–442 (2017). doi:10.1126/science.aam9321
- Ch. 23 — Chen J.S., Ma E., Harrington L.B., et al. CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity. Science 360, 436–439 (2018). doi:10.1126/science.aar6245
- Ch. 23 — Contagion Live. FDA authorizes first CRISPR test kit for COVID-19 (May 2020).
- Ch. 24 — FDA. 510(k) K082562, Applied Biosystems 7500 Fast Dx (decision 30 Sep 2008).
- Ch. 24 — FDA. 510(k) records for open real-time instruments: K033734, LightCycler Instrument version 1.2 (17 Dec 2003); K091409, LightCycler MRSA Advanced Test for use with LightCycler 2.0 (6 Jul 2010); K113319, Rotor-Gene Q MDx (6 Feb 2012); K123955, QuantStudio Dx Real-Time PCR Instrument (8 Mar 2013). openFDA 510(k) data (api.fda.gov/device/510k.json), accessed 17 Oct 2026.
- Ch. 24 — FDA. Establishment registration and device listing: QuantStudio 5 Dx and QuantStudio 7 Pro Dx (Life Technologies Holdings Pte Ltd, product code OOI); CFX Opus 96 Dx, CFX Opus 384 Dx, CFX Opus Deepwell Dx, CFX96 Dx and CFX96 Deep Well Dx (Bio-Rad Laboratories (S) Pte Ltd, product code NSU). openFDA registration and listing data (api.fda.gov/device/registrationlisting.json), accessed 17 Oct 2026.
- Ch. 24 — 21 CFR 862.2570, Instrumentation for clinical multiplex test systems (eCFR); FDA product classification database, product code OOI.
- Ch. 24 — Roche. Roche launches next-generation qPCR system (LightCycler PRO), investor update (20 Nov 2023). roche.com/investors/updates/inv-update-2023-11-20 (company statement; also the source for TIB Molbiol's LightMix assay counts)
- Ch. 24 — FDA. 510(k) records for digital PCR tests: K181661, Bio-Rad QXDx BCR-ABL %IS Kit on the QXDx AutoDG ddPCR System (decision 13 Feb 2019), 510(k) summary at accessdata.fda.gov/cdrh_docs/pdf18/K181661.pdf; K221869, Suzhou Sniper Medical Technologies, BCR-ABL1 (p210) % IS Kit (Digital PCR Method) (5 Sep 2023); K252243, Suzhou Precigenome, Fastplex BCR-ABL (p210) %IS digital PCR Kit (2 Jul 2026). openFDA 510(k) data, device-name search, accessed 17 Oct 2026.
- Ch. 24 — QIAGEN. QIAGEN launches QIAcuityDx digital PCR system for clinical testing in oncology (press release, GlobeNewswire, 30 Sep 2024). globenewswire.com (company statement)
- Ch. 24 — Bio-Rad. Frequently asked questions: Bio-Rad's acquisition of Stilla Technologies (web page, 2025). bio-rad.com (company statement)
- Ch. 24 — Bio-Rad. QX700 E Droplet Digital PCR System, Bulletin 3826 (July 2025). bio-rad.com/sites/default/files/2025-07/Bulletin_3826.pdf (manufacturer)
- Ch. 24 — FDA. 510(k) records for central-laboratory systems: Roche cobas CT/NG for the cobas 6800/8800 Systems, K173887 (21 Mar 2018), and for the cobas 5800/6800/8800 Systems, K260917 (8 Jul 2026); Hologic Panther Fusion Flu A/B/RSV, K171963 (26 Sep 2017), GI Bacterial, K251868, and GI Expanded Bacterial, K251993 (both 25 Sep 2025); Abbott Alinity m STI, K202977 (29 Apr 2022), and HCV, K252102 (25 Sep 2025). openFDA 510(k) data, accessed 17 Oct 2026.
- Ch. 24 — BD. Diagnostic Solutions – Medical Professionals (web page, 2026). bd.com (company statement)
- Ch. 24 — FDA. 510(k) records cleared as Dual Track (510(k) with CLIA waiver by application): Cepheid Xpert Flu+RSV Xpress, K151226 (3 Dec 2015), to Xpert Hemorrhagic Fever, K261362 (7 Aug 2026); cobas Influenza A/B & RSV for cobas Liat (applicant Iquum, Inc.), K153544 (25 Jul 2016), to cobas liat CT/NG and CT/NG/MG, K253756 and K253759 (15 May 2026); BioFire Spotfire Respiratory Panel, K213954 (3 Feb 2023), and Respiratory/Sore Throat Panel Mini, K243544 (14 Aug 2025). openFDA 510(k) data, accessed 17 Oct 2026.
- Ch. 24 — 42 CFR 493.17, Test categorization (eCFR); FDA, CLIA categorizations (web page).
- Ch. 24 — FDA. 510(k) records for QIAstat-Dx panels (traditional 510(k)s): Respiratory Panel, K183597 (18 May 2019), to BCID GPF Plus AMR, K254194 (6 Aug 2026), and BCID GN Plus AMR, K262168 (24 Sep 2026). openFDA 510(k) data, accessed 17 Oct 2026.
- Ch. 24 — FDA. De Novo decision summary DEN240020, Visby Medical Women's Sexual Health Test (28 Mar 2025). accessdata.fda.gov/cdrh_docs/reviews/DEN240020.pdf
- Ch. 24 — Visby Medical. Visby Medical receives landmark FDA De Novo authorization for first-ever OTC PCR test for sexual health (press release, 2025). poc.visbymedical.com (company statement)
- Ch. 24 — FDA. 510(k) records for Visby Medical tests: K200748 (26 Aug 2021), K220407 (7 Mar 2023), K242526 (19 Feb 2025) and K251501 (1 Oct 2025), Dual Track; K253971, Flu and COVID-19 Test (27 May 2026), traditional. openFDA 510(k) data, accessed 17 Oct 2026.
- Ch. 24 — FDA. De Novo decision summary DEN220028, Cue COVID-19 Molecular Test (6 Jun 2023). accessdata.fda.gov/cdrh_docs/reviews/DEN220028.pdf
- Ch. 24 — Becker's Hospital Review. Cue Health files for bankruptcy (29 May 2024).
- Ch. 24 — FDA. Revocation of the emergency use authorizations of the Cue COVID-19 Test and the Cue COVID-19 Test for Home and Over The Counter Use, effective 9 Oct 2024. Federal Register 89(224) (20 Nov 2024), FR doc 2024-27094.
- Ch. 24 — GenomeWeb. Pfizer buys Lucira Health for $36.4M through bankruptcy auction (April 2023).
- Ch. 24 — 360Dx. Report on Pfizer ending the Lucira by Pfizer business (2025; page undated). 360dx.com/node/99851
- Ch. 24 — FDA. Revocation of the emergency use authorizations of the Lucira by Pfizer COVID-19 & Flu tests, effective 22 Oct 2025. Federal Register public inspection document 2025-24155.
- Ch. 24 — Danaher. Our businesses (web page, accessed 17 Oct 2026). danaher.com/our-businesses (company statement)
- Ch. 24 — New England Biolabs. About NEB (web page, accessed 17 Oct 2026). neb.com/en/about-neb (company statement)
- Ch. 24 — Takara Bio USA. About (web page, accessed 17 Oct 2026). takarabio.com/about (company statement)
- Ch. 24 — ICMR, FIND and Molbio Diagnostics. WHO endorses Truenat for the initial diagnosis of TB and detection of rifampicin resistance (joint press release, 2 Jul 2020). icmr.gov.in/icmrobject/custom_data/1702893145_pr_truenat_who_endorsement_02072020.pdf
- Ch. 24 — WHO Prequalification. Mycobacterium tuberculosis first and/or second line anti-TB drugs tests (web page, accessed 17 Oct 2026). extranet.who.int/prequal/key-resources/documents/mycobacterium-tuberculosis-first-andor-second-line-anti-tb-drugs-tests
- Ch. 24 — ICMR. Guidelines for validation and batch testing of COVID-19 diagnostic kits (4 Jun 2020).
- Ch. 24 — ICMR. Performance evaluation of commercial kits for real time PCR for COVID by ICMR identified validation centres (17 Feb 2022), copy hosted by PRS Legislative Research. prsindia.org/files/covid19/notifications/RT_PCR_Tests_Kits_Evaluation_Summ_17022022.pdf
- Ch. 24 — Co-Diagnostics. Co-Diagnostics JV CoSara first Indian company to receive license for COVID-19 diagnostic (press release, Business Wire, 19 Mar 2020). businesswire.com (company statement)
- Ch. 24 — BioSpectrum India. CDSCO approves COVID-19 PCR kit by Mylab (24 Mar 2020). biospectrumindia.com (company claim as reported)
- Ch. 24 — Co-Diagnostics. Co-Diagnostics joint venture CoSara receives CDSCO license to manufacture and sell CoSara PCR Pro instrument (press release via BioSpace, 5 Feb 2026). biospace.com (company statement)
- Ch. 24 — CDSCO. List of in-vitro diagnostic medical devices under Medical Devices Rules, 2017 (updated 25 Oct 2023).
- Ch. 24 — Government of India. Medical Devices Rules, 2017, G.S.R. 78(E) (31 Jan 2017).
- Ch. 24 — QIAGEN. QIAGEN to discontinue NeuMoDx integrated PCR testing system, support customers during transition period (press release, 6 Jun 2024). corporate.qiagen.com (company statement)
- Ch. 24 — Roche. Media release on completing the acquisition of LumiraDx's point-of-care technology (29 Jul 2024). roche.com/media/releases/med-cor-2024-07-29 (company statement)
- Ch. 24 — CHEManager. Bio-Rad acquires Stilla Technologies (8 Jul 2025). chemanager-online.com
- Ch. 24 — Waters Corporation. Release on completing the combination with BD's Biosciences and Diagnostic Solutions business (PR Newswire, 9 Feb 2026); BD, news release on the BD–Waters transaction (2026) (company statements)
- Ch. 24 — Hologic. Form 8-K, exhibit 99.1 (SEC EDGAR, 2026), and release, Blackstone and TPG complete acquisition of Hologic (7 Apr 2026) (company statements)
- Ch. 24 — Molbio Diagnostics. Press release under Regulation 30, NSE filing (7 Sep 2026) (company statement); ICICI Securities, Molbio Diagnostics IPO review (Aug 2026).
- Ch. 25 — CDC. Laboratory advisory: COVID emergency declaration ended and impact to emergency use authorized (EUA) tests (14 Oct 2026). cdc.gov/locs/php/messages/10-14-2026-laboratory-advisory-covid-emergency-declaration-ended-and-impact-to-emergency-use.html
- Ch. 25 — FDA. Transition Plan for Medical Devices Issued Emergency Use Authorizations Related to COVID-19 (final guidance, March 2023).
- Ch. 25 — FDA. Medical Devices; Exemptions From Premarket Notification for Class I and Class II Devices. Final order, 84 FR 71794 (30 Dec 2019).
- Ch. 25 — 21 CFR Part 820, Quality Management System Regulation (eCFR), §820.7 and §820.10.
- Ch. 25 — FDA. Revocation of authorization of emergency use of an in vitro diagnostic device for detection and/or diagnosis of COVID-19; availability. 86 FR 28841 (28 May 2021), recording De Novo DEN200031, granted 17 Mar 2021; classification codified at 89 FR 66554 (16 Aug 2024).
- Ch. 25 — CLP Magazine. FDA permits marketing of first COVID-19 diagnostic using the De Novo review pathway (18 Mar 2021). Trade press.
- Ch. 25 — FDA. 510(k) database: K231481, Cepheid Xpert Xpress CoV-2/Flu/RSV plus (17 Aug 2023), product code QOF.
- Ch. 25 — 21 U.S.C. 360bbb-3 (Federal Food, Drug, and Cosmetic Act section 564).
- Ch. 25 — FDA. openFDA 510(k) dataset, clearance type "Dual Track" (dual 510(k) and CLIA waiver by application). api.fda.gov/device/510k.json (FDA labels openFDA data unvalidated).
- Ch. 25 — FDA. Medical Devices; Laboratory Developed Tests. Final rule, 89 FR 37286 (6 May 2024); Laboratory Developed Tests; Implementation of Vacatur, 90 FR 45134 (19 Sep 2025); Covington & Burling, FDA's LDT rule struck down by federal district court (3 Apr 2025).
- Ch. 25 — 21 CFR 860.3, Definitions: class I, class II and class III, the three categories of regulatory control (eCFR); FDA, Classify your medical device (web page, content current as of 26 Aug 2026).
- Ch. 26 — MDCG 2020-16. Guidance on classification rules for in vitro diagnostic medical devices under Regulation (EU) 2017/746, rev. 4 (March 2025) and rev. 5 (September 2026).
- Ch. 26 — Team-NB. Considerations for conformity assessments done by (IVD) notified bodies in the case of down classification of SARS-CoV-2 tests (12 Dec 2025).
- Ch. 26 — Commission Implementing Regulation (EU) 2022/1107 laying down common specifications for certain class D in vitro diagnostic medical devices. OJ L 178, 5.7.2022.
- Ch. 26 — European Commission. EU reference laboratories (in vitro diagnostics web page).
- Ch. 26 — Regulation (EU) 2024/1860 amending Regulations (EU) 2017/745 and (EU) 2017/746 (13 Jun 2024); HPRA, IVDR transitional provisions (web page); IGJ, Transitional periods for the MDR and IVDR (web page).
- Ch. 26 — European Commission. COM(2025) 1023 final (16 Dec 2025), with its explanatory memorandum; European Parliament Legislative Observatory, procedure 2025/0404(COD) (seen 17 Oct 2026).
- Ch. 26 — Agence Europe. Medical devices: member states back goal to reach general approach before end of Irish Presidency of EU Council (15 Oct 2026). News service.
- Ch. 26 — Lok Sabha. Unstarred Question No. 1669, answered 16 Dec 2022 (Ministry of Health and Family Welfare).
- Ch. 26 — Government of India. G.S.R. 777(E) (14 Oct 2022), exemption of non-sterile and non-measuring class A medical devices from licensing, as listed on CDSCO's Medical Devices Rules page; CDSCO, Risk based classification list of class A (non-sterile and non-measuring) medical devices (31 Oct 2025).
- Ch. 26 — CDSCO. In-vitro diagnostic medical devices: frequently asked questions, CDSCO/IVD/FAQ/04/2022, with addendum of 28 Mar 2025 (FAQ 44).
- Ch. 26 — CDSCO. In-vitro diagnostic medical devices: frequently asked questions, CDSCO/IVD/FAQ/03/2022.
- Ch. 26 — ICMR. Standard protocol for validation of commercial RT-PCR based SARS-CoV-2 detection kit (undated; copy hosted by RGCB).
- Ch. 26 — ICMR. Notice inviting comments on draft standard evaluation protocols for IVDs (31 Dec 2024).
- Ch. 26 — CDSCO. Brief note on the release of the ICMR–CDSCO compendium of standard evaluation protocols for IVDs (11 Sep 2025); ICMR and CDSCO, Compendium of standard IVD evaluation protocols (2025).
- Ch. 26 — CDC. Telebriefing transcripts, 5 Feb 2020 and 28 Feb 2020 (not edited by CDC). archive.cdc.gov
Take the PDF with you.
Reading online needs no sign-up. For the PDF edition, laid out for print and offline reading, tell us where to send it and we will email you.
Check your inbox.
The download link is on its way to your email. If it has not arrived in a few minutes, check your spam folder or write to info@unplex.tech.
Preview only — nothing was sent
Let's Talk