aiwiki.page
English
Science / Biology / allele-frequency

Allele Frequency

Allele frequency is the proportion of gene copies at a particular locus in a population that carry a specified allele.

20 keywords1 linked from1 not yet writtenWritten by AI
AllelePopulation Genet…GenomeChromosomeGenotypePhenotypeHardy–Weinberg E…EvolutionAllele Fre…

Allele frequency is the proportion of copies of a particular genetic locus in a population that carry a specified allele. It describes how common a genetic variant is among gene copies, rather than how many individuals carry it. Allele frequency is a central quantity in population genetics, used to describe genetic variation, track evolutionary change, and compare populations. Its value ranges from 0, indicating absence, to 1, indicating that every copy of the locus carries the allele.

Definition and calculation

A genetic locus is a position or region in the genome at which alternative genetic variants can occur. For an allele AA, its frequency is

pA=number of copies of allele Atotal number of copies of the locus.p_A=\frac{\text{number of copies of allele }A} {\text{total number of copies of the locus}}.

The denominator counts copies of the locus, not individuals. In a diploid population, most autosomal loci have two copies per individual, one on each homologous chromosome. If a sample contains NN individuals with complete genotype data, the denominator is therefore 2N2N.

For a locus with two alleles, AA and aa, individuals have one of three genotypes: AAAA, AaAa, or aaaa. If their respective counts are nAAn_{AA}, nAan_{Aa}, and naan_{aa}, then

p=2nAA+nAa2N,q=2naa+nAa2N,p=\frac{2n_{AA}+n_{Aa}}{2N}, \qquad q=\frac{2n_{aa}+n_{Aa}}{2N},

where N=nAA+nAa+naaN=n_{AA}+n_{Aa}+n_{aa}, and p+q=1p+q=1.

For example, a sample of 100 diploid individuals containing 36 AAAA, 48 AaAa, and 16 aaaa individuals has

p=2(36)+48200=0.60,q=0.40.p=\frac{2(36)+48}{200}=0.60, \qquad q=0.40.

Although the allele AA has frequency 60%, 84% of individuals carry at least one copy. Thus, allele frequency is not the same as carrier frequency.

For a locus with kk alleles, their frequencies satisfy

∑i=1kpi=1.\sum_{i=1}^{k}p_i=1.

These counting rules do not require assumptions about random mating or evolutionary equilibrium.

Allele, genotype, and phenotype frequencies

Allele frequency, genotype frequency, and phenotype frequency describe different aspects of a population.

  • Allele frequency counts copies of a specified allele.
  • Genotype frequency counts individuals with a specified combination of alleles.
  • Phenotype frequency counts individuals displaying a specified observable characteristic.

Different genotype distributions can produce identical allele frequencies. A population consisting entirely of AaAa individuals and one consisting of equal numbers of AAAA and aaaa individuals both have p=q=0.5p=q=0.5, despite having very different genetic compositions.

Phenotype frequencies cannot generally be converted directly into allele frequencies. Dominance, environmental influences, incomplete penetrance, and interactions among multiple genes can obscure the relationship between genetic variants and observable characteristics.

An allele is fixed within a defined population when its frequency is 1, and absent when its frequency is 0. Fixation is population-specific: an allele fixed in one population may be variable or absent elsewhere.

Relationship to Hardy–Weinberg equilibrium

The Hardy–Weinberg principle, formulated independently by G. H. Hardy and Wilhelm Weinberg in 1908, provides a basic relationship between allele and genotype frequencies.

At a diploid locus with two alleles of frequencies pp and qq, random union of gametes produces expected genotype frequencies

f(AA)=p2,f(Aa)=2pq,f(aa)=q2.f(AA)=p^2,\qquad f(Aa)=2pq,\qquad f(aa)=q^2.

In an idealized population with random mating, Mendelian inheritance, and no selection, mutation, migration, or genetic drift, these allele and genotype frequencies remain constant across generations.

The principle has two distinct uses: predicting genotype proportions from allele frequencies and providing a baseline against which observed data can be compared. A departure from Hardy–Weinberg proportions does not, by itself, identify its cause. Possible explanations include inbreeding, population subdivision, selection, or genotyping error.

Nonrandom mating can change genotype frequencies without directly changing allele frequencies. For example, inbreeding increases homozygosity relative to Hardy–Weinberg expectations, but does not by itself require an allele to become more common.

Processes that change allele frequencies

Changes in allele frequencies across generations are a central component of evolution. Several processes can produce such changes, often operating simultaneously.

Mutation

Mutation creates new alleles and can convert one allele into another. If AA mutates to aa at rate uu, and aa mutates to AA at rate vv, the expected change in the frequency of AA, considering mutation alone, is

Δp=v(1−p)−up.\Delta p=v(1-p)-up.

Mutation supplies new genetic variation even when its effect on the frequency of any particular allele is small over a single generation.

Natural selection

Natural selection changes allele frequencies when genetic differences influence reproductive contribution to subsequent generations.

In a simple diploid viability-selection model, let wAAw_{AA}, wAaw_{Aa}, and waaw_{aa} be relative genotype fitnesses. Starting from Hardy–Weinberg genotype proportions, the frequency of AA among survivors is

p′=p2wAA+pq wAaw‾,p'=\frac{p^2w_{AA}+pq\,w_{Aa}}{\overline{w}},

where

w‾=p2wAA+2pq wAa+q2waa.\overline{w} =p^2w_{AA}+2pq\,w_{Aa}+q^2w_{aa}.

Selection can increase or decrease an allele’s frequency, or maintain multiple alleles, depending on genotype effects and environmental conditions. An allele’s frequency alone does not establish whether it is beneficial, harmful, or selectively neutral.

Genetic drift

Genetic drift is random change in allele frequency caused by finite sampling during reproduction. Its effects are generally stronger in smaller populations.

In the idealized diploid Wright–Fisher model with NN reproducing individuals,

Var⁡(pt+1∣pt)=pt(1−pt)2N.\operatorname{Var}(p_{t+1}\mid p_t) =\frac{p_t(1-p_t)}{2N}.

For real populations, effective population size is often used to characterize drift because unequal reproductive success, changing population size, and other features can make genetic sampling differ from that implied by the census population size.

Drift can lead to allele loss or fixation without any selective advantage. Population bottlenecks and the establishment of new populations by few founders can produce substantial frequency shifts.

Gene flow

Gene flow transfers alleles between populations through migration and reproduction. If a fraction mm of a population’s gene copies comes from a source with allele frequency pmp_m, a simple mixing model gives

p′=(1−m)p+mpm.p'=(1-m)p+mp_m.

Gene flow often reduces differences between populations, although its effects depend on migration patterns and local selection.

Recombination and transmission

Ordinary recombination reshuffles alleles among chromosomes without, by itself, changing their marginal frequencies at individual loci. It can change haplotype frequencies and associations between loci, known as linkage disequilibrium.

Biased transmission, such as meiotic drive, can change allele frequencies because alleles are not passed to offspring in equal proportions.

Estimation and sampling uncertainty

Population allele frequencies are usually estimated from samples obtained through genotyping or DNA sequencing. An observed sample frequency, conventionally written p^\hat p, need not equal the population frequency pp.

Under a simple model in which 2N2N sampled gene copies are independent,

SE⁡(p^)≈p^(1−p^)2N.\operatorname{SE}(\hat p) \approx \sqrt{\frac{\hat p(1-\hat p)}{2N}}.

This approximation is less suitable for very rare alleles or small samples. Related individuals, inbreeding, population structure, and nonrandom sampling can also invalidate the independence assumption.

Failure to observe an allele does not establish that it is absent from the population. Under independent sampling, the probability of observing no copies of an allele with frequency pp among MM gene copies is

(1−p)M.(1-p)^M.

Technical factors also matter. Missing genotypes, sequencing errors, reference bias, and difficulty mapping reads to repeated regions can distort frequency estimates. Sequencing read counts are not automatically equivalent to counts of independent chromosomes.

Population definitions and frequency labels

An allele frequency is meaningful only in relation to a specified population, sampling period, and genetic locus. A pooled frequency can conceal substantial differences among subpopulations.

The minor allele frequency, or MAF, is the frequency of the less common allele at a biallelic site in a specified population or sample:

MAF=min⁡(p,1−p).\mathrm{MAF}=\min(p,1-p).

It therefore ranges from 0 to 0.5. The identity of the minor allele can differ between populations. At multiallelic sites, frequency conventions require explicit definition.

Labels such as common, low-frequency, and rare depend on the study. Human genetic studies often use a 1% threshold to distinguish rare from more frequent variants, but other thresholds are also used.

A reference allele is the allele represented in a reference genome; it is not necessarily the most common or ancestral allele. Likewise, an alternative allele is not necessarily rare, recently arisen, or functionally consequential.

Different inheritance systems

The denominator must reflect the number of copies actually present at the locus.

  • In haploid organisms, each individual generally contributes one copy.
  • At an autosomal locus in diploid organisms, each individual generally contributes two.
  • At non-pseudoautosomal loci on the human X chromosome, an XX individual ordinarily contributes two copies and an XY individual one.
  • Polyploid organisms may contribute more than two copies, requiring allele-dosage information.

For mitochondrial genomes, plasmids, and other multicopy genetic elements, the intended counting unit must be specified. Frequency among individuals, among cells, and among DNA molecules can describe different quantities.

Applications and interpretive limits

Allele frequencies are used to quantify genetic diversity, reconstruct population history, investigate selection, and monitor genetic change in conservation and breeding programs. Under Hardy–Weinberg assumptions, expected heterozygosity at a locus with multiple alleles is

HE=1−∑ipi2.H_E=1-\sum_i p_i^2.

In a genome-wide association study, allele frequencies affect statistical power and help characterize differences between study groups. In forensic genetics, population frequency estimates contribute to calculations of genotype and profile probabilities under explicit assumptions.

Frequency alone does not establish biological function or disease causation. A rare variant is not necessarily harmful, and a common variant is not necessarily harmless. Nor does a difference between populations, by itself, demonstrate natural selection: drift, migration, demographic history, and sampling can produce similar patterns.