Genetic Distance Measures: Review

Abstract

The major unsolved problem of descriptive population genetics is an adequate specification of the genetic difference between two closely related species as compared to the genetic difference between two populations of the same species. Traditional population genetic analyses deal with the distribution of allele frequencies between and within populations. From these frequencies several measures of population structure can be estimated, one of the most widely used being the genetic distance measures. Genetic distance is the degree of gene difference (genomic difference) between species or populations that is measured by some numerical method. Genetic distance measures have already been established as one of the major tools for analyzing data on gene differentiation between populations. Many genetic distances have been developed, of which a few remain in regular use. Each of these genetic distances has unique evolutionary and statistical properties, and evolutionary relationships inferred from each genetic distance can be quite different. Quantification of the genetic distance between populations is instrumental in many genetic research initiatives, and a large number of formulas for this purpose have been proposed. However, selection of an appropriate measure for assessing genetic distance between real-world human populations that diverged as a result of mechanisms that are not fully known can be a challenging task. In this study twenty six distance measures were investigated. For macroevolutionary comparisons, Nei's measures are probably the best. In microevolutionary studies, when sample sizes are approximately equal and the differences in gene frequency are great, Edward's E 2 is preferable. If sample sizes are quite variable and gene frequencies do not differ greatly, Sanghvi's G2 would be most appropriate.