hardy-weinberg-genetics

Hardy-Weinberg equilibrium math for a single biallelic locus, plus a chi-square goodness-of-fit test to check whether an observed population departs from equilibrium. Pure ESM, zero dependencies.

Model

For alleles A and a with frequencies p and q (p + q = 1), the equilibrium genotype frequencies are:

AA = p^2      Aa = 2pq      aa = q^2

Genotype counts are passed as { AA, Aa, aa }.

Usage

import { alleleFrequencies, hweTest } from 'hardy-weinberg-genetics';

alleleFrequencies({ AA: 298, Aa: 489, aa: 213 });
// { p: 0.5425, q: 0.4575 }

const r = hweTest({ AA: 0, Aa: 100, aa: 0 });
// r.chi2, r.df (=1), r.pValue, r.expected, r.inEquilibrium

API

• alleleFrequencies({AA,Aa,aa}) → { p, q }
• expectedFrequencies(p) → { AA, Aa, aa }
• expectedCounts(p, n) → expected genotype counts
• chiSquare(observed) → { chi2, df, expected }
• chiSquarePValue(chi2, df) → upper-tail p-value
• hweTest(observed, alpha=0.05) → full report incl. inEquilibrium

The p-value uses a series expansion of the regularized lower incomplete gamma function with a Lanczos ln-gamma; it's accurate to a few parts in 10^-4 across the usual chi-square range, which is well past what a genotype table needs.

Note on df

A biallelic HWE test has 1 degree of freedom: three genotype categories, minus one for the fixed total, minus one for the estimated allele frequency.

License

BSD-3-Clause