A drop of blood or a single hair holds only a tiny amount of DNA, far too little to study. The polymerase chain reaction (PCR) makes billions of copies of one chosen piece. It is DNA replication done in a tube, over and over.
Key idea
PCR repeats a cycle of three temperatures. Denature at about 95 °C: the strands separate. Anneal at about 50–65 °C: short primers bind on each side of the target. Extend at about 72 °C: a heat-proof DNA polymerase builds new strands. Each cycle doubles the target, so n cycles make $2^n$ copies of each starting copy.
Why it works
Why doubling? Every copy made in one cycle becomes a template in the next. One copy becomes 2, then 4, then 8. That is exponential growth: after n cycles there are $2^n$ times as many copies. Thirty cycles give $2^{30}$, about a billion, in a couple of hours.
Worked example: Counting copies
A tube starts with 5 copies of a gene. How many copies are there after 10 cycles?
- Each cycle doubles: 10 cycles multiply by $2^{10}$, which is 1,024.
- 5 × 1,024 = 5,120.
Answer: 5,120 copies (in an ideal reaction).
Worked example: How many cycles?
Starting from one copy, how many cycles are needed to pass one million copies?
- $2^{19}$ is 524,288, which is not enough.
- $2^{20}$ is 1,048,576, which passes one million.
Answer: 20 cycles.
Two discoveries made PCR practical. Kary Mullis invented the method in 1983, and shared the 1993 Nobel Prize in Chemistry for it. And the DNA polymerase used comes from Thermus aquaticus, a bacterium from hot springs; this "Taq" polymerase survives the 95 °C step, so it does not have to be added again every cycle.
- Primers are short single strands of DNA, about 20 bases, made to match the two ends of the target.
- Only the stretch between the two primers is copied over and over.
- Real reactions slow down after many cycles as primers and nucleotides run low, so $2^n$ is the ideal.
Common mistake
Two slips are common. One is adding instead of multiplying: 10 cycles do not make 10 or 20 copies, they make $2^{10}$ = 1,024 times as many. The other is thinking PCR copies a whole genome. It copies only the piece between the two primers. Check: the number of copies should double each time you add one cycle.