Counting the share of a population that shows a trait is simple division. The real skill is predicting that share from gene versions you can’t see. Hardy-Weinberg is the tool most problems hinge on, because it links the visible traits to the hidden gene combinations behind them. Its neat equations describe an idealized population, though, and real groups that inbreed or are very small drift away from the predictions.
The Basic Phenotype Frequency Formula
Three different things get counted in these problems. A phenotype is the trait you can see, like flower color or eye color. A genotype is the pair of gene versions behind it, such as AA, Aa or aa. An allele is one gene version, and each individual carries two for most genes.
Phenotype frequency is a share of individuals: the number with the trait divided by the total number of individuals. If 30 of 200 plants have white flowers, the frequency is 30 ÷ 200 = 0.15, or 15%. Every phenotype frequency in a population adds up to 1, or 100%.
Allele frequency works differently because it counts gene copies, not individuals. If you know genotype counts, p = (2×AA + Aa) ÷ (2×N) and q = (2×aa + Aa) ÷ (2×N), where N is the number of individuals.1Pearson. Allele Frequency Calculator | Find p and q + Hardy–Weinberg Check Each individual carries two copies, so a population of N has 2N alleles.
Take 100 individuals: 36 AA, 48 Aa and 16 aa. The dominant phenotype frequency is (36 + 48) ÷ 100 = 0.84. For the alleles, p = (72 + 48) ÷ 200 = 0.6 and q = (32 + 48) ÷ 200 = 0.4. Check that p + q = 1. The most common mistake is dividing allele counts by N instead of 2N, which makes the allele frequencies add to more than 1.
Using Hardy-Weinberg to Predict Phenotype Frequencies
Hardy-Weinberg equilibrium is the state where allele and genotype frequencies stay constant from generation to generation. It starts with p + q = 1 for the two allele frequencies. Genotype frequencies then follow p² + 2pq + q² = 1, for AA, Aa and aa. That works because, under random mating, an offspring’s two alleles are like two independent random draws from the population’s gene pool.
The equation is what you get by multiplying (p + q) by (p + q), which mirrors picking one allele from each parent. The 2 in 2pq exists because a heterozygous individual (one with two different alleles, Aa) can form two ways: A from the mother and a from the father, or the reverse. Leaving out the 2 is a common error that makes the totals fall short of 1.
Only the homozygous recessive genotype (two copies of the recessive allele, aa) shows the recessive trait, so that trait’s frequency is q². Both AA and the carrier (Aa) show the dominant trait, so its frequency is p² + 2pq, which is also simply 1 − q².
Suppose q = 0.2. The recessive trait appears in 0.04, or 4%. Then p = 0.8, so AA is 0.64 and Aa is 0.32. The dominant phenotype is 0.96, matching 1 − 0.04. Carriers outnumber affected individuals eight to one here.
That ratio explains why rare recessive traits hide. When q is small, q² is tiny but 2pq is much larger, so most recessive gene versions sit in people who look normal.
To check whether a population fits Hardy-Weinberg, compute p and q from the observed genotype counts, then multiply p², 2pq and q² by N to get expected counts. Close agreement with the observed counts is consistent with Hardy-Weinberg, while a large gap suggests one of its assumptions, such as random mating, is not being met. Small samples show some mismatch naturally, so courses usually use a chi-square test to decide whether a gap is too big to be chance.
Working Backward From the Recessive Phenotype
Start by finding which phenotype belongs to the two recessive copies (aa).2Albert. Hardy-Weinberg Equation: AP® Biology Crash Course Its frequency directly gives q².3Albert. Hardy-Weinberg Equation: AP® Biology Crash Course It is the only genotype you can identify for sure from appearance, since AA and Aa look the same.4Albert. Hardy-Weinberg Equation: AP® Biology Crash Course Occasionally a problem names the trait but not which allele is dominant, and then you need the stated inheritance pattern before deciding which phenotype is q².
A classic practice problem gives aa as 36% of a population and asks for every allele and genotype frequency.5Kansas State University. Hardy-Weinberg practice questions Here q² = 0.36, so q = √0.36 = 0.6. Then p = 1 − 0.6 = 0.4. AA = 0.4² = 0.16 and Aa = 2 × 0.4 × 0.6 = 0.48. The dominant phenotype is 0.16 + 0.48 = 0.64, which with 0.36 makes 1.
The method only works from the recessive phenotype because the dominant one is a mixture of AA and Aa. You can’t read p² or 2pq from it alone, and only the recessive class maps to a single genotype. So never take the square root of the dominant percentage. If you’re given the dominant share, subtract it from 1 first to get q², then take the square root.
For any exam problem, the steps run like this:
- Identify the recessive phenotype and convert its percent to a decimal.
- Take the square root to get q, then subtract from 1 for p.
- Compute p², 2pq and q², and confirm they sum to 1.
- Answer exactly what was asked: allele, genotype or phenotype.
That last step matters for carriers. Once you have 2pq, multiply by the population size to get a head count: 0.48 carriers in 1,000 individuals is 480 people. Counts of individuals should be whole numbers, while frequencies are decimals, so notice whether the question wants a frequency, a percentage or a number of people.
Blood Types and Other Traits With Three Alleles
For ABO, the frequencies of the A, B and O alleles add to 1.6Bioinformation. Phenotype, allele and genotype frequency distribution of ABO and Rh(D) blood group among blood donors attending regional blood transfusion centre in Delhi, India The genotypes follow A² + 2AO + B² + 2BO + 2AB + O² = 1.7Bioinformation. Phenotype, allele and genotype frequency distribution of ABO and Rh(D) blood group among blood donors attending regional blood transfusion centre in Delhi, India That gives type A frequency = A² + 2AO, type B = B² + 2BO, type AB = 2AB, and type O = O².8Bioinformation. Phenotype, allele and genotype frequency distribution of ABO and Rh(D) blood group among blood donors attending regional blood transfusion centre in Delhi, India
There are six genotypes because every pairing of two copies is possible: AA, BB and OO, plus AB, AO and BO. The mixed pairings each carry a 2 for the same reason carriers do in the two-allele case, since either parent can supply either allele. Four phenotypes come from these six genotypes, because AA and AO both give type A and BB and BO both give type B.
The O allele frequency can be estimated as the square root of the share of people with type O.9Let’s Talk Academy. How to Calculate ABO Blood Group Allele Frequencies from Population Data O is recessive to both A and B, so type O is the one fully known genotype. A and B are each dominant over O, and they are codominant with each other, so a person with both shows type AB rather than a blend or one winning out. Because O is hidden inside two other phenotypes, the A and B classes can’t be turned into allele frequencies directly the way type O can.
Call the allele frequencies p for A, q for B and r for O. After finding r, note that type O plus type A equals (p + r)². So p = √(O + A) − r, and then q = 1 − p − r. As a check, the predicted type AB should match 2pq.
X-Linked Traits Need Separate Math for Males and Females
In fruit flies, a female with one red-eye allele has red eyes even if she also carries white.10Science Primer. X-Linked Inheritance So females can be XWXW or XWXw with red eyes, or XwXw with white eyes.11Science Primer. X-Linked Inheritance Males have only one X, so they are either XWY with red eyes or XwY with white eyes.12Science Primer. X-Linked Inheritance
That changes the math. A male’s phenotype shows whichever allele he carries, with no dominance masking anything, so the male frequency of a recessive trait equals q directly. Females have two X chromosomes, so ordinary Hardy-Weinberg applies and their recessive frequency is q².
If q = 0.1, 10% of males show the trait, but only 1% of females do, since they need two copies. This is why X-linked recessive traits such as red-green color blindness are mostly seen in men.
The allele frequencies p and q describe all X chromosomes in the gene pool, and the same numbers serve both sexes. In a Hardy-Weinberg population, female genotypes are p², 2pq and q², while male genotypes are simply p and q. A mixed sample requires keeping the sexes separate when counting.
Working backward is easier with male data: count affected males to read q off directly, with no square root. Then use that q to predict female carriers, 2pq.
How Inbreeding Shifts Phenotype Frequencies
The inbreeding coefficient, F, is the chance that an individual’s two alleles are identical copies inherited from one shared ancestor.13UBC Zoology. How to Calculate Phenotype Frequency in a Population With chance F the second allele must match the first; otherwise it is drawn at random.14UBC Zoology. How to Calculate Phenotype Frequency in a Population So the chance of being A1A1 is p²(1−F) + pF.15UBC Zoology. How to Calculate Phenotype Frequency in a Population
Heterozygotes become 2pq(1−F) and aa becomes q²(1−F) + qF.16Case Studies in Ecology and Evolution. Inbreeding and Population Structure At F = 1, complete inbreeding, there are no heterozygotes at all.17Case Studies in Ecology and Evolution. Inbreeding and Population Structure Inbreeding changes how alleles pair into genotypes, not how many copies of each allele exist, so p and q stay put. The extra homozygotes come at the expense of heterozygotes, which shrink by the factor (1 − F). That is why inbreeding raises the recessive phenotype frequency even though the allele frequency is unchanged.
Take q = 0.1, so random mating gives 1% recessive. With F = 0.25, aa = 0.01 × 0.75 + 0.1 × 0.25 = 0.0325. The recessive trait frequency more than triples, which is why rare recessive conditions cluster in closely related families. In data, the signal is fewer heterozygotes than 2pq predicts, and F = 0 returns the ordinary Hardy-Weinberg equations.
Why Small Populations and Small Samples Mislead
By chance alone, genetic drift can make populations lose genetic variation.18Understanding Evolution. Effects of genetic drift All populations experience drift, but the smaller the population, the sooner it has a drastic effect.19Understanding Evolution. Effects of genetic drift20National Institute of Justice. Population Genetics and Statistics for Forensic Analysts | Genetic Drift and Natural Selection
Each generation’s alleles are a random sample of the last generation’s, and small samples stray further from the true proportions. A rare allele in a tiny group can be lost entirely, and once lost it can’t return without migration or mutation. In a large population, random ups and downs mostly cancel out. Drift doesn’t favor better-adapted individuals, so it can raise or lower a harmful allele’s frequency.
Samples have their own limits. To be 95% sure an allele frequency estimate is within 0.05 of the truth, a sample of roughly 49 to 285 individuals is needed.21PubMed Central. Confidence Intervals for Population Allele Frequencies: The General Case of Sampling from a Finite Diploid Population of Any Size Being within 0.1 needs only about 22 to 94 individuals.22PubMed Central. Confidence Intervals for Population Allele Frequencies: The General Case of Sampling from a Finite Diploid Population of Any Size A class sample of 20 therefore gives only a rough estimate.
When a Genotype Doesn’t Show Up as the Expected Trait
In large population studies, the share of carriers of single-gene disease variants who develop the condition averages 60% or lower for most conditions.23PubMed Central. Incomplete Penetrance and Variable Expressivity: From Clinical Studies to Population Cohorts Geneticists call this share penetrance. About 1 in 75 healthy elderly people carried a known disease variant, such as for inherited high cholesterol, without having the disease.24PubMed Central. Incomplete Penetrance and Variable Expressivity: From Clinical Studies to Population Cohorts Other genes and factors can protect people from a variant’s effect.25PubMed Central. Incomplete Penetrance and Variable Expressivity: From Clinical Studies to Population Cohorts
Textbook problems assume everyone with the genotype shows the trait. To adjust the calculation, multiply the predicted genotype frequency by the share of carriers who show the trait. For example, q² = 0.04 with 60% showing it gives 2.4% affected.
Dominance and penetrance are separate ideas. Dominance describes which allele shows when two different ones are present, while penetrance describes whether a genotype shows the trait at all. A person can have a dominant disease allele and still be healthy if penetrance is incomplete, which doesn’t contradict the inheritance pattern.
The same idea applies to family trees. A dominant trait that seems to skip a generation may reflect incomplete penetrance rather than a recessive pattern. The person who skipped carries the variant, as shown by an affected parent and affected children, but never developed the trait. Pedigree rules that assume full penetrance can misclassify some families.