Standard Article

Stationary Allele Frequency Distributions

  1. Bruce Rannala

Published Online: 15 APR 2013

DOI: 10.1002/9780470015902.a0005465.pub3



How to Cite

Rannala, B. 2013. Stationary Allele Frequency Distributions. eLS. .

Author Information

  1. University of California, Davis, California, USA

Publication History

  1. Published Online: 15 APR 2013


Forces that determine the allele frequencies in natural populations include genetic drift, natural selection, migration and mutation. A balance of opposing forces can, in some cases, cause allele frequencies to approach a stationary distribution over time. The form of this distribution is not influenced by initial allele frequencies, but instead is determined by the relative magnitudes of different evolutionary forces. Statistical distributions are presented for the stationary allele frequencies under several simple population genetic models (including the k alleles symmetrical mutation model and the Wright island model of migration). In addition, the sampling distribution of allele counts under these models are described. The latter is useful when using genetic marker data to estimate population parameters. A population is at genetic equilibrium if evolutionary forces have persisted long enough for a population to have reached the stationary distribution, this is not often the case in nature.

Key Concepts:

  • If evolutionary forces such as natural selection, genetic drift and migration remain constant allele frequencies may reach to a stationary distribution.

  • Because evolutionary forces are often changing on a relatively short time-scale the allele frequencies in many populations will not be at equilibrium.

  • Statistics based on parameters of stationary distributions may provide a useful summary of the genetic variation in populations that are not at equilibrium.

  • Simulation methods can be used to study allele frequency distributions for more complex models.

  • Modern approaches focus on nonstationary sampling distributions derived using the coalescent process model.


  • probability distribution;
  • Fisher-Wright model;
  • genetic drift;
  • migration;
  • mutation;
  • sampling distribution