Hardy-Weinberg Equilibrium

The Hardy-Weinberg equation is a fundamental principle in population genetics that predicts allele and genotype frequencies in non-evolving populations using the formula: p² + 2pq + q² = 1, where p represents the frequency of the dominant allele (A) and q represents the frequency of the recessive allele (a), with p + q = 1. The genotype frequencies are: p² = frequency of homozygous dominant (AA), 2pq = frequency of heterozygous (Aa), and q² = frequency of homozygous recessive (aa). For example, if a population has 16% of individuals with blue eyes (a recessive trait, genotype aa), then q² = 0.16, so q = √0.16 = 0.4 (40% recessive allele frequency). Since p + q = 1, then p = 1 - 0.4 = 0.6 (60% dominant allele frequency). The genotype frequencies would be: AA = p² = (0.6)² = 0.36 (36%), Aa = 2pq = 2(0.6)(0.4) = 0.48 (48%), and aa = q² = 0.16 (16%). This means in a population of 1,000 individuals, you'd expect approximately 360 homozygous dominant, 480 heterozygous carriers, and 160 homozygous recessive individuals. The equation applies when five conditions are met: large population size, random mating, no mutations, no gene flow (migration), and no natural selection affecting the trait.

Hardy-Weinberg equilibrium requires five specific conditions that, when violated, cause allele frequencies to change across generations: (1) Large population size—small populations (typically <1,000 individuals) experience genetic drift where random sampling causes allele frequency changes of ±5-10% per generation. For example, in a population of 50 individuals, random chance alone can shift an allele from 50% to 45% or 55% frequency. (2) Random mating—non-random mating patterns change genotype frequencies without affecting allele frequencies. Inbreeding increases homozygosity by 10-25%, creating more AA and aa individuals but fewer Aa heterozygotes than predicted. Assortative mating (like-with-like) has similar effects. (3) No mutations—mutation rates typically range 1×10⁻⁸ to 1×10⁻⁶ per base pair per generation in humans. While low, over evolutionary time these introduce new alleles. A mutation rate of 1×10⁻⁵ changes allele frequency by 0.001% per generation. (4) No gene flow—migration introduces new alleles or changes frequencies. Even 1-5% migration per generation substantially alters allele frequencies; for instance, if population A (p=0.7) receives 10% migrants from population B (p=0.3), the new frequency becomes p = 0.9(0.7) + 0.1(0.3) = 0.66. (5) No selection—natural selection is the most powerful force changing allele frequencies. A fitness disadvantage of just 1% (selection coefficient s=0.01) causes significant frequency changes; a lethal recessive allele (s=1) decreases from q=0.1 to q=0.09 in one generation. In reality, few populations meet all five conditions, but Hardy-Weinberg provides a null hypothesis to detect evolution—deviations from expected frequencies indicate evolutionary forces are acting on the population.

Hardy-Weinberg calculations are clinically important for estimating carrier frequencies of recessive genetic diseases, which is essential for genetic counseling and screening programs. The key insight is that most recessive disease alleles exist in heterozygous carriers (Aa) who are phenotypically normal but can pass the disease allele to offspring. For example, cystic fibrosis (CF) affects approximately 1 in 3,500 Caucasians (q² = 1/3,500 = 0.000286). Solving for q: q = √0.000286 = 0.017 (1.7% disease allele frequency). The carrier frequency is 2pq = 2(0.983)(0.017) = 0.033 or approximately 1 in 30 Caucasians. This means for every person with CF, there are about 115 healthy carriers. Sickle cell disease in African Americans has incidence of 1 in 365 (q² = 0.00274), giving q = 0.052 and carrier frequency 2pq = 0.099 or about 1 in 10. Tay-Sachs disease in Ashkenazi Jews has incidence 1 in 3,600 (q² = 0.000278), giving carrier frequency of 1 in 30. These calculations inform genetic screening recommendations and help predict offspring risk. If both parents are carriers (Aa × Aa), each pregnancy has a 25% chance (¼) of affected child (aa), 50% chance (½) of carrier child (Aa), and 25% chance (¼) of non-carrier child (AA). If carrier frequency is 1/30, then random couple has (1/30)² = 1/900 chance both are carriers, and conditional probability of affected child = (1/900) × (1/4) = 1/3,600, matching the disease frequency. This validates the Hardy-Weinberg prediction and enables population-level risk assessment for genetic counseling programs.

Allele frequencies and genotype frequencies are distinct but mathematically related concepts that have different clinical implications. Allele frequency represents the proportion of a specific allele in the population's total gene pool, calculated as the number of copies of that allele divided by total alleles (2× population size for diploid organisms). Genotype frequency represents the proportion of individuals with a specific genotype combination. For a trait with alleles A (frequency p = 0.7) and a (frequency q = 0.3), the allele frequencies sum to 1.0 (p + q = 1), but genotype frequencies are: AA = p² = 0.49 (49%), Aa = 2pq = 0.42 (42%), and aa = q² = 0.09 (9%), which also sum to 1.0. Clinically, this distinction matters for disease risk prediction. Consider familial hypercholesterolemia (FH), where the disease allele frequency is approximately q = 0.002 (0.2%). The disease occurs in homozygotes (aa) at frequency q² = 0.000004 (1 in 250,000), causing severe early heart disease. However, heterozygous carriers (Aa) occur at frequency 2pq = 2(0.998)(0.002) = 0.00399 (1 in 250), who have moderately elevated cholesterol and 3-fold increased cardiovascular risk requiring treatment. This means carriers are 1,000 times more common than affected individuals. For hemochromatosis (HFE C282Y mutation), the allele frequency in Northern Europeans is q ≈ 0.065, giving homozygote frequency q² = 0.0042 (1 in 238) and carrier frequency 2pq = 0.122 (1 in 8). While 1 in 238 have the genotype, only 10-20% develop clinical disease due to incomplete penetrance, demonstrating that genotype frequency doesn't always equal disease frequency. Understanding this relationship is essential for interpreting genetic test results and counseling patients about disease risk versus carrier status.

Testing for Hardy-Weinberg equilibrium involves comparing observed genotype frequencies in a population to expected frequencies calculated from allele frequencies, typically using the chi-square (χ²) goodness-of-fit test. The procedure: (1) Count observed genotypes in the population sample. For example, in 1,000 individuals: 640 AA, 320 Aa, 40 aa. (2) Calculate observed allele frequencies: p = [2(640) + 320] / [2(1,000)] = 1,600/2,000 = 0.8; q = 1 - p = 0.2. (3) Calculate expected genotype frequencies using Hardy-Weinberg: AA expected = p² × 1,000 = (0.8)² × 1,000 = 640; Aa expected = 2pq × 1,000 = 2(0.8)(0.2) × 1,000 = 320; aa expected = q² × 1,000 = (0.2)² × 1,000 = 40. (4) Calculate χ² = Σ[(observed - expected)² / expected] across all genotypes. (5) Compare to critical χ² value with degrees of freedom = (number of genotypes - number of alleles) = 3 - 2 = 1. At p < 0.05, critical χ² = 3.841. If calculated χ² > 3.841, reject Hardy-Weinberg equilibrium. Deviations from equilibrium indicate evolutionary forces: Excess homozygotes (more AA and aa, fewer Aa than expected) suggests inbreeding, population substructure, or null alleles (alleles that fail to amplify in genetic testing). For example, if observed counts were 680 AA, 240 Aa, 80 aa (vs. expected 640, 320, 40), this pattern indicates inbreeding coefficient F ≈ 0.25. Deficit of homozygotes (fewer AA and aa, more Aa) suggests heterozygote advantage (overdominance), as seen with sickle cell trait in malaria-endemic regions where heterozygotes (HbAS) have fitness advantage. Deficit of one homozygote class suggests selection against that genotype or problems with sampling/genetic testing. In medical genetics, Hardy-Weinberg testing validates genetic data quality—failure to meet equilibrium may indicate genotyping errors, population stratification in case-control studies, or true biological phenomena like selection or non-random mating that affect disease risk calculations.

Hardy-Weinberg equilibrium applies differently to X-linked traits because males have only one X chromosome (hemizygous), while females have two (diploid), creating distinct frequency patterns between sexes. For an X-linked recessive trait with disease allele frequency q, males are affected at frequency q (they only need one copy), while females are affected at frequency q² (they need two copies). The ratio of affected males to affected females is q : q² = 1 : q, meaning males are affected 1/q times more frequently. For example, red-green color blindness (X-linked recessive) has allele frequency q ≈ 0.08 in Caucasian populations. Affected males = q = 0.08 (8%, or 1 in 12.5), while affected females = q² = 0.0064 (0.64%, or 1 in 156). Males are 12.5 times more likely to be color blind. Carrier females = 2pq = 2(0.92)(0.08) = 0.147 (14.7%, nearly 1 in 7 women). Hemophilia A has allele frequency q ≈ 0.0001, affecting 1 in 10,000 males (q = 0.0001) but only 1 in 100,000,000 females (q² = 0.00000001), making affected females extremely rare. Female carriers occur at 2pq ≈ 0.0002 (1 in 5,000). Duchenne muscular dystrophy (DMD) has q ≈ 0.0001, affecting 1 in 10,000 males; affected females are virtually non-existent (1 in 100 million), though carrier females may show mild symptoms due to X-inactivation (lyonization) where random inactivation of one X chromosome can cause mosaic expression. In genetic counseling, if a woman is a carrier (Xᴺ Xᵈ) and her partner is unaffected (Xᴺ Y), each son has 50% chance of being affected (Xᵈ Y) and each daughter has 50% chance of being a carrier (Xᴺ Xᵈ). The skewed sex ratio in X-linked diseases makes Hardy-Weinberg calculations essential for carrier screening and family planning decisions.