Use Play to run a full PCR cycle, or Step to advance one stage at a time. Drag to rotate the scene and click any structure — template strand, primer, polymerase, new strand, or a free nucleotide — to learn what it does. Drag the Cycles slider to see the copies multiply.
Often you have only a vanishingly small amount of DNA — a smear at a crime scene, a few virus particles in a swab, one gene you want to study. PCR is the photocopier of molecular biology: it takes one chosen stretch of DNA and makes billions of identical copies in a couple of hours, so there is finally enough to detect or work with. It was invented by Kary Mullis in 1983 and earned him the 1993 Nobel Prize in Chemistry.
PCR is just heating and cooling a tube on a schedule. Each cycle has three temperature steps. Denaturation at about $94^\circ$C breaks the hydrogen bonds so the double helix splits into two single strands. Annealing at about $55^\circ$C lets two short primers stick to the ends of the target, one on each strand. Extension at $72^\circ$C is when Taq polymerase adds free nucleotides onto each primer, building a fresh complementary strand. One double helix has become two.
Because every cycle copies every strand present, the number of target molecules doubles each time: $1\to 2\to 4\to 8\to\dots$ After $n$ cycles you have about $2^{n}$ copies. After $30$ cycles that is $2^{30}\approx 1.07\times 10^{9}$ — over a billion — from a single starting molecule. Two design tricks make this clean: the two primers flank exactly the region you want, so only that piece (the amplicon) is amplified; and Taq, from a hot-spring bacterium, survives the $94^\circ$C step so the same enzyme works cycle after cycle without re-adding it.
Press Play to run denature → anneal → extend and watch the strands split, the primers dock, and a new green strand grow. Then drag the Cycles run slider from $1$ toward $30$ and watch the Copies readout explode — that is $2^{n}$ in action. Toggle Primers off to see why polymerase has nothing to start from, and turn Free nucleotides off to see the raw building blocks disappear.
| Stage | Temperature | What happens |
|---|---|---|
| 1. Reaction mix | ~25 °C | Template, primers, Taq, dNTPs and Mg²⁺ buffer combined |
| 2. Denature | ~94 °C | Hydrogen bonds break; the helix separates into single strands |
| 3. Anneal | ~55 °C | Forward and reverse primers bind their matching sequences |
| 4. Extend | ~72 °C | Taq polymerase adds nucleotides 5′→3′, building new strands |
| 5. Cycle doubles | cycle end | One double helix has become two identical copies |
| 6. Repeat ×n | ~25–35 cycles | Exponential: ≈2ⁿ copies (30 cycles ≈ 1 billion) |
If amplification were perfect, $C$ copies after $n$ cycles would be $C = C_0\,2^{n}$, where $C_0$ is the starting number of target molecules. Starting from one molecule: $10$ cycles gives about a thousand ($2^{10}=1024$), $20$ gives about a million, and $30$ gives over a billion. In practice efficiency is a little below $100\%$ and reagents eventually run low, so growth slows and then plateaus — which is exactly the curve quantitative PCR (qPCR) measures to tell how much DNA was there to start with.
In the very first cycle the new strands run past the far primer site, so they are longer than the target. But from the second cycle on, strands made off those primers are themselves copied starting at the opposite primer, producing pieces bounded by both primers — the fixed-length amplicon. Within a few cycles these short, primer-to-primer products vastly outnumber everything else, so the tube fills with copies of exactly the region the scientist chose.
PCR (the polymerase chain reaction) is a lab technique that makes millions to billions of copies of a specific stretch of DNA from a tiny starting sample. It is used everywhere in biology and medicine: diagnosing infections (including COVID-19 tests), DNA fingerprinting in forensics, paternity testing, detecting genetic disorders, and cloning genes for research.
Key takeaway: PCR copies a chosen DNA segment millions of times, so a tiny sample becomes enough to detect or analyse.Each cycle is defined by temperature. (1) Denaturation at about $94$–$96^\circ$C: heat breaks the hydrogen bonds and the helix separates into single strands. (2) Annealing at about $50$–$65^\circ$C: the mixture cools so short primers stick to their matching sequences. (3) Extension at about $72^\circ$C: the polymerase adds nucleotides to each primer, building a new complementary strand. Then it repeats.
Key takeaway: denature (hot, split strands), anneal (cool, primers bind), extend (72°C, polymerase copies) — repeated over and over.Primers are short, single-stranded pieces of DNA (about $18$–$25$ bases) designed to match the sequences at the two ends of the target region. There are two — a forward and a reverse primer — one for each strand. They mark exactly where copying starts and stops, so PCR amplifies only the chosen segment. DNA polymerase cannot begin a strand from scratch; it can only add onto an existing primer.
Key takeaway: the two primers flank the target and tell the polymerase where to start, so only the chosen region is amplified.PCR repeatedly heats the mixture to about $94^\circ$C to separate the strands, and that heat would destroy a normal DNA polymerase. Taq polymerase comes from Thermus aquaticus, a bacterium that lives in hot springs, so it survives the high temperatures and works best near $72^\circ$C. Using Taq means you do not have to add fresh enzyme every cycle, which is what let PCR be automated in a thermal cycler.
Key takeaway: Taq polymerase, from a hot-spring bacterium, survives the 94°C step, so the same enzyme works through every cycle.Every cycle copies each piece of target DNA, so the amount doubles: $1\to2\to4\to8$, and so on. After $n$ cycles you have about $2^{n}$ copies. After $30$ cycles that is roughly $2^{30}$, or over a billion copies, from a single starting molecule. This doubling-each-cycle pattern is what exponential growth means.
Key takeaway: each cycle doubles the DNA, so after n cycles there are about 2ⁿ copies — 30 cycles gives over a billion.After the first couple of cycles, most new strands run only from one primer site to the other, because new strands made from primers are themselves copied starting at the other primer. The result is a product of fixed length bounded by the two primers, called the amplicon. By choosing where the primers bind, a scientist sets exactly which region — and how long a piece — gets amplified.
Key takeaway: the two primers define a fixed-length product called the amplicon; choosing the primers chooses the region copied.