Dihybrid Cross Punnett Square Calculator
Predict two-gene offspring genotypes and dominant/recessive phenotype probabilities from two parental genotypes.
Parent genotypes
The model assumes complete dominance and independent assortment. Uppercase alleles are treated as dominant; lowercase alleles are recessive.
Live results
Phenotype distribution
Phenotype probabilities
| Phenotype class | Count | Probability |
|---|
Punnett square
How to use this dihybrid cross calculator
What this calculator does
This calculator predicts the possible offspring from a cross involving two genes. It constructs each parent's possible gametes, combines them in a Punnett square, and summarizes genotype and phenotype probabilities. It is useful for classroom genetics, checking hand calculations, preparing breeding examples, and exploring how homozygous or heterozygous parents alter expected outcomes. It does not predict an individual offspring with certainty, model linked genes, account for incomplete dominance, or replace empirical breeding data.
How to calculate
- Enter the Parent 1 genotype as four letters representing two allele pairs, such as AaBb.
- Enter the Parent 2 genotype using the same two gene letters and the same gene order.
- Read the live summary cards, the phenotype distribution, and the complete Punnett square. No Calculate button is needed because results update as you type.
- Use Reset to restore the default AaBb × AaBb cross. Use Download Excel to save the current inputs, summary, phenotype probabilities, and full square as a validated workbook.
Input guide
Parent 1 genotype and Parent 2 genotype are required four-character text values. Each must contain two alleles for the first gene and two for the second gene. Uppercase means dominant and lowercase means recessive. Valid examples include AaBb, AABb, aaBB, and aabb. The parents must use the same gene letters; entering AaBb for one parent and CcDd for the other is rejected. A common mistake is mixing the gene order, such as writing AbBa, or using more than two different letters.
Output guide
Both dominant traits is the probability that an offspring has at least one dominant allele for each gene. Gene 1 dominant and Gene 2 dominant show the marginal probability of expressing each dominant trait. Double recessive is the probability of inheriting two recessive alleles at both genes. The phenotype table divides all outcomes into four mutually exclusive classes, while the Punnett square shows every genotype combination. The header pills report each parent's distinct gamete count and the total number of equally likely outcome cells.
Worked example
For AaBb × AaBb, each parent makes four equally likely gametes: AB, Ab, aB, and ab. Combining 4 gametes by 4 gametes produces 16 cells. Nine cells contain at least one A and one B, so the probability of both dominant traits is 9/16 = 56.25%. Three cells are dominant for gene 1 but recessive for gene 2, three are recessive for gene 1 but dominant for gene 2, and one is double recessive. This gives the classic 9:3:3:1 phenotype ratio.
Learn more
The model follows Mendel's law of independent assortment, explained in the OpenStax overview of inheritance laws. For a guided explanation of the 9:3:3:1 pattern, see Khan Academy's dihybrid cross lesson.
How the model works
Each gamete receives one allele from each gene. The calculator enumerates those gametes directly, preserving duplicate probabilities when a parent is homozygous at a locus. Every maternal-paternal gamete pair is then combined into a canonical genotype. Phenotypes are classified by the presence or absence of at least one uppercase allele at each gene. This probability method is exact for the simple Mendelian model and agrees with the multiplication rule described in the OpenStax Concepts of Biology inheritance chapter.
Independent assortment is an assumption, not a universal biological rule. Genes that are physically linked can deviate from the four equally likely gametes expected from a double heterozygote. Dominance relationships can also differ from the complete-dominance convention used here. In those cases, a linkage map, recombination frequency, or a different phenotype model is needed.