Punnett Square Calculator

How to use this calculator

1. Write each parent as allele pairsTwo letters for a single gene, four for two genes. Use the same letter in upper and lower case for the two alleles of one gene — Aa, not Ab.
2. Keep the gene order consistentIf parent one is AaBb, parent two must also list the A gene first. Writing BbAa for the second parent will produce a nonsense result.
3. Choose the dominance modelComplete dominance means one allele masks the other entirely. Incomplete or co-dominance means the heterozygote shows its own distinct phenotype, which changes the phenotype ratio but never the genotype ratio.
4. Read both ratiosThe genotype ratio counts the actual allele combinations. The phenotype ratio counts what you would observe. They are different, and exam questions usually want a specific one.

How the calculation works

A Punnett square is a systematic way of listing every possible fertilisation. Its logic comes from meiosis: each parent contributes exactly one allele per gene to each gamete, and which one is a matter of chance.

For a single gene, a heterozygote Aa produces two kinds of gamete, A and a, in equal proportion. Crossing two heterozygotes gives a 2 × 2 grid with four cells: AA, Aa, aA and aa. Since Aa and aA are the same genotype, the genotype ratio is 1 AA : 2 Aa : 1 aa.

Under complete dominance, AA and Aa look identical, so the phenotype ratio collapses to 3 showing the dominant trait to 1 showing the recessive. That 3:1 is the single most recognisable result in genetics and it comes straight out of the grid.

For two genes the number of gametes doubles per gene. AaBb produces four gamete types — AB, Ab, aB and ab — giving a 4 × 4 grid with sixteen cells. Crossing two AaBb parents produces the classic 9:3:3:1 phenotype ratio, which is what independent assortment looks like when you count it.

The rule generalises: n heterozygous genes give 2ⁿ gamete types and 4ⁿ grid cells.

The standard ratios

The Aa × aa cross is worth knowing separately — it is the test cross, used to determine whether an individual showing a dominant phenotype is homozygous or heterozygous. Cross it with a recessive: if any offspring show the recessive trait, the unknown parent must have carried a recessive allele.

Worked example: a dihybrid cross

Cross AaBb with AaBb. Each parent produces four gamete types: AB, Ab, aB and ab.

That gives a 4 × 4 grid with sixteen cells. Counting the phenotypes under complete dominance, nine offspring show both dominant traits, three show dominant A with recessive b, three show recessive a with dominant B, and one shows both recessives. The 9:3:3:1 ratio.

The genotype tally is more granular: 1 AABB, 2 AABb, 2 AaBB, 4 AaBb, 1 AAbb, 2 Aabb, 1 aaBB, 2 aaBb and 1 aabb. Sixteen cells, nine distinct genotypes.

Notice that AaBb appears four times — a quarter of all offspring. Double heterozygotes are the most common single genotype in this cross precisely because there are four different ways to assemble one.

What the square does not model

A Punnett square assumes independent assortment, which holds only when the genes sit on different chromosomes or far apart on the same one. Genes close together on the same chromosome are linked and inherited together far more often than the grid predicts. That is genuine biology the square cannot represent, and it is why real dihybrid data frequently deviates from 9:3:3:1.

It also assumes simple Mendelian dominance. Many real traits are polygenic, involving many genes each contributing a small effect, which produces continuous variation rather than discrete categories. Height and skin colour behave this way, and no Punnett square describes them.

Epistasis is another complication: one gene masking the expression of another entirely, which distorts the phenotype ratio without changing the genotype ratio at all.

And the fundamental caveat: the grid gives probabilities, not outcomes. A 3:1 ratio means each offspring has a 3-in-4 chance of the dominant phenotype, independently. Four offspring from that cross can perfectly well all show the recessive trait, and frequently do in a small family. The ratio emerges over large numbers, which is why Mendel counted thousands of plants rather than dozens.

Assumptions and limitations

• Assumes independent assortment — linked genes on the same chromosome do not follow these ratios
• Handles one or two genes only; three or more produce grids too large to be useful
• Does not model sex-linked inheritance, which needs X and Y rather than plain alleles
• Does not handle epistasis, polygenic traits or multiple alleles beyond a simple pair
• Gives probabilities per offspring, not guaranteed outcomes in a small family
• Assumes both parents contribute equally and that all genotypes are equally viable

Questions people ask about this in Philippines