In Silico PCR: How to Predict a PCR Product Before You Run It
6 min read · Updated July 10, 2026
Ordering a primer pair and finding out on a gel that it also amplifies a second locus, or amplifies nothing at all, wastes a week and a tube of reagent. In silico PCR simulates the reaction against a template or a genome before you order anything, reporting where each primer binds, what size product forms, and whether more than one product is possible. This guide covers how the simulation works, what it catches that a quick sequence search misses, and how to read the output when checking primers for cloning or genotyping.
What the simulation actually does
In silico PCR takes a template sequence — a plasmid, a genomic region, or a full genome — and two primer sequences, then works out where a real PCR reaction would produce a product. It scans the sense strand for a match to the forward primer and scans the reverse complement of the same template for a match to the reverse primer. When a forward match is followed downstream by a reverse match, it reports the amplicon between them: the sequence, its length, and its start and end coordinates on the template.
The size calculation is simple once both binding sites are known: it's the number of bases from the first base of the forward primer's match to the last base of the reverse primer's match, inclusive. If the forward primer binds starting at position 120 and the reverse primer's match ends at position 540, the predicted product is 421 bp. Knowing this number before you run a gel means you know exactly what band to expect, and any other band means something unexpected annealed.
Why the 3' end decides whether a product forms
A DNA polymerase can only extend a primer from its 3' end, so that end has to pair correctly with the template for extension to start at all. A single mismatch at or near the 3' terminal base is usually enough to stop amplification outright, even if the rest of the primer matches well. A specificity search that allows mismatches should treat a 3'-end mismatch as effectively disqualifying a site, rather than averaging it into an overall similarity score.
The 5' end behaves differently. Extra bases tacked onto the 5' end — a restriction site for cloning, a tag, a sequencing adapter — don't need to match the template at all, and real PCR tolerates a few mismatches near the 5' end of an otherwise well-matched primer. A tool that flags every 5' mismatch as a broken primer will miss primer designs that work fine in the tube; one that's too loose about 3' mismatches will predict products that never actually form.
The main problem it solves: specificity before you order
The single most useful thing in silico PCR catches is off-target binding. A primer pair can look fine on paper — reasonable Tm, no obvious hairpins — and still match a second site in the template or genome. If it does, the reaction produces more than one product: a genomic PCR might throw a faint second band, or a wrong-size band might outcompete the one you wanted. Checking the primers in silico first means every predicted product gets reported, not just the one you were hoping for, so a multi-band or wrong-size result is something you catch on screen instead of on a gel three days later. A specificity check against a template or genome should tell you:
- how many places the primer pair produces a product
- the predicted size of each product
- where each product sits relative to the feature you care about
Where this fits in a normal primer workflow
In practice, in silico PCR usually shows up at one of four points in a project: verifying a primer pair is specific to the intended template before ordering it, predicting the expected band size so a gel result can be checked against a number rather than a guess, designing genotyping or screening primers meant to distinguish a wild-type allele from a knockout or an insertion (where the point is that the two alleles give different products), and confirming that primers spanning a cloning junction amplify at the size the construct predicts, which is a fast way to catch a misassembled clone before sequencing it.
What it won't tell you
In silico PCR is a geometric simulation: it finds where primers match and reports the product that geometry implies. It does not model secondary structure in the template, polymerase-specific behavior such as processivity or strand displacement, or the effect of PCR additives like betaine or DMSO. A predicted product can exist on paper and still amplify poorly in a real tube if, for example, the template folds into a stable structure at one of the primer sites. Treat the prediction as what the primers should produce given clean annealing, not as a guarantee of yield or efficiency.
Checking a primer pair with SeqBench's In-silico PCR tool
SeqBench's In-silico PCR tool runs this check directly: enter a template and two primers, and it predicts the product, its size, and its position in the template. Use it to confirm a pair is specific before you place an order, to know the band size to expect on a gel, or to sanity-check a genotyping pair against both a wild-type and a modified template. Once primers are in hand, Primer Tm can confirm the melting temperature (Tm) and GC% for the reaction, and Primer Designer can generate a new ranked pair from scratch if the one you tested doesn't hold up.
Frequently asked questions
How does in silico PCR predict the size of a PCR product?
It finds where the forward primer matches the sense strand and where the reverse primer matches the reverse complement, then reports the product as the stretch of template between those two binding sites, inclusive. The predicted size is just the base count of that stretch, which you can compare directly to what shows up on a gel.
Can in silico PCR tell me if my primers will bind more than one place?
Yes, that is its main use before ordering primers. If a primer pair matches more than one site in the template or genome, it reports a product for each match, so you can see a multi-band result coming instead of finding it on a gel.
Do mismatches at the 5' end of a primer stop it from working in PCR?
Usually not. Extension only starts from the 3' end, so PCR tolerates 5' mismatches, such as an added restriction site, far better than 3' mismatches, which can block extension outright even when the rest of the primer matches well.
Does a predicted PCR product mean the reaction will definitely work in the lab?
Not entirely. In silico PCR simulates primer binding and extension geometrically, so it doesn't account for template secondary structure, polymerase behavior, or PCR additives; it tells you what product the primer positions imply, not how efficiently the real reaction will run.