Punnett Square Calculator

See how parental alleles become gametes and offspring combinations. Compare genotype and phenotype groups, highlight outcomes meeting your conditions, and explore why an expected count is not a guaranteed result in a finite sample.

Genetics probability explorer

Follow the alleles. Understand the odds.

Explore an entered educational cross, trace gamete combinations, and see how a probability differs from a guaranteed sample result.

Chance per offspring

6.25%

Exact: 1/16

Expected matches

1

Across 16 independent offspring; not a guaranteed count.

At least one match

64.393%

Across 16 independent offspring.

Parents & target conditions

One to three independent loci, two alleles each. A/a, B/b and C/c identify the entered alleles; choose dominance separately. Target conditions are combined with AND. No linkage, sex linkage, penetrance or real-trait prediction is modeled.

Locus A

An illustrative label, not a verified trait or gene.

Locus B

An illustrative label, not a verified trait or gene.

Enter a whole number from 1 to 100 independent offspring. Expected counts are averages over repeated samples.

Target probability 6.25%. 9 possible genotypes and 4 phenotype groups.

Gametes meet in the Punnett square

Parent 1 is on the left; parent 2 is across the top. Identical gametes are combined. Each cell gives its genotype and exact probability. “Match” identifies cells meeting every target condition.

4 by 4 unique gamete combinations
Parent 1 ↓ / Parent 2 →AB1/4Ab1/4aB1/4ab1/4
AB1/4
Ab1/4
aB1/4
ab1/4

Selected gamete pair

AB × AB → AA BB

1/4 × 1/4 = 1/16. Does not meet all target conditions.

Outcome probabilities

Select an outcome to highlight all contributing cells. Bars share a fixed 0–100% scale.

A probability is not a fixed count

Distribution of matches in 16 independent offspring. The horizontal axis is the number of matches, from 0 to 16. Bar height is probability, scaled to the most likely count. Use the selector or complete table for exact values.

0 matches: 35.607%

Complete sample distribution & exact fractions
MatchesProbabilityExact fraction
035.607%6568408355712890625/18446744073709551616
137.981%437893890380859375/1152921504606846976
218.991%437893890380859375/2305843009213693952
35.9082%68116827392578125/1152921504606846976
41.2801%59034583740234375/4611686018427387904
50.20482%2361383349609375/1152921504606846976
60.025033%577227041015625/2305843009213693952
70.0023841%27487001953125/1152921504606846976
80.00017881%16492201171875/9223372036854775808
91.060e-5%122164453125/1152921504606846976
104.945e-7%11402015625/2305843009213693952
111.798e-8%207309375/1152921504606846976
124.995e-10%23034375/4611686018427387904
131.025e-11%118125/1152921504606846976
141.464e-13%3375/2305843009213693952
151.301e-15%15/1152921504606846976
165.421e-18%1/18446744073709551616

Keep the complete cross

Export authored assumptions, all gametes and cells, grouped outcomes, target probabilities and every sample-count probability. Exact fractions are preserved.

How to use this calculator

  1. Choose one to three independent loci. Enter each parent’s genotype and an illustrative label for each locus.
  2. Choose the phenotype grouping at each locus. Uppercase and lowercase identify alleles here; dominance is separately selected.
  3. Set a target condition for each locus. A matching offspring must meet every condition, and Any genotype leaves that locus unrestricted.
  4. Select a Punnett-square cell to trace its parental gametes and probability. Switch outcome views and select a group to highlight all contributing cells.
  5. Enter an illustrative sample size from 1 to 100. Inspect the distribution of matching counts, then download the complete CSV or use the copyable report.

The Formula

The mathematical logic powering this calculation is:

For each independent locus, each parent contributes one of its two alleles with equal transmission probability. Identical gametes are combined. A grid cell has probability P(parent 1 gamete) × P(parent 2 gamete). Add cell probabilities to obtain a genotype or an entered phenotype group. Target conditions are joined with AND across loci. The exact transmission denominator is 4^L for L loci; the model supports L = 1, 2 or 3.
Phenotype grouping is an explicit assumption: first allele dominant, second allele dominant, or three distinguishable genotype groups. Changing grouping does not change genotype probabilities. Linkage, sex linkage, penetrance, epistasis, mutation and viability differences are not modeled.
For n independent offspring with the same selected-event probability p, P(K=k) = C(n,k) p^k (1−p)^(n−k), E[K] = np, and P(K≥1) = 1−(1−p)^n. The sample limit is 1–100. Integer arithmetic retains exact fractions; displayed percentages are rounded and do not turn a nonzero probability into zero or a near-certain event into certainty.

Worked Example

If you input:

  • Parents: AaBb × AaBb; two independent loci
  • Grouping: First allele dominant at both loci
  • Target: aa at locus A AND bb at locus B
  • Sample: 16 independent offspring

Result:

The four gametes AB, Ab, aB and ab each have probability 1/4 for each parent. The target aabb occupies one of 16 equally likely cells, so its probability is 1/16 = 6.25%. The phenotype groups have ratio 9:3:3:1. In 16 independent offspring the expected target count is 1, but the chance of at least one is 1 − (15/16)^16, about 64.39%; exactly one match is not guaranteed.

Frequently Asked Questions

Why does my square have fewer rows than the textbook example?

Identical gametes are combined. A homozygous parent contributes only one allele type at that locus. The headers show each unique gamete’s exact probability, and a cell multiplies its two header probabilities.

Does uppercase always mean dominant here?

No. A/a, B/b and C/c identify the two entered allele versions at each locus. The separate phenotype-grouping control specifies which allele is dominant, or whether all three genotypes remain distinguishable.

What does the three-group option mean?

It keeps the two homozygous genotypes and the heterozygote in separate phenotype groups. It does not infer a physical intermediate, a particular codominant trait or a biological mechanism from the letters you enter.

Can I use this for linked genes or sex-linked inheritance?

This model assumes independent biallelic loci with equal segregation. It does not accept linkage phase, recombination fractions or sex-chromosome rules. Those situations need a different model.

Why is the expected count different from the most likely count?

Expected count is a probability-weighted average over repeated samples and can be fractional. The distribution shows the probability of each whole count in a sample. Neither a mean nor a most-likely count guarantees what one sample contains.

Do all target conditions have to occur together?

Yes. The target is an AND across loci. Any genotype means that locus imposes no restriction. At least one first or second allele includes the heterozygote as well as the corresponding homozygote.

Can this predict a real person’s traits or medical risks?

The workspace is an educational model of the entered cross and grouping assumptions. It does not establish real genotypes, gene–trait relationships, penetrance or clinical risk. Do not interpret its illustrative outputs as a prediction about a person.

Similar Calculators