Statistics of microsatellite loci: estimation of mutation rate and pattern of population expansion

Abstract

Abstract Applications of microsatellite loci in evolutionary studies are based on examination of the expectations of summary statistics of genetic variation at these loci, under a model for mutation and genetic drift. Usually, the single-step stepwise mutation model with unrestricted allelic states is assumed, and the effect of genetic drift is studied by assuming that the population size is constant over time. In this chapter, we illustrate the effects of relaxing both of these assumptions. We show that an expansion or contraction of population size over time leaves a signature on the genetic variation that can be detected by estimating the composite parameter (the product of population size and the mutation rate) from heterozygosity and allele size variance. Allele size restrictions also produce an imbalance between heterozygosity and allele size variance. These two statistics are insensitive to directional (expansion/contraction) bias of mutations if allele-size constraints are absent. Further, the feature that microsatellite alleles can be ordered by their sizes allows estimation of relative mutation rates of loci grouped by their biological properties. Applications of these principles to microsatellite data from several human populations indicate that the mutation rates are inversely related to the motif size, and the African populations exhibit indications of the most ancient human bottleneck.