Statistical measures of genetic differentiation of populations: Rationales, history and current states

Population differentiation is a fundamental process of evolution, and many evolutionary studies, such as population genetics, phylogeography and conservation biology, all require the inference of population differentiation. Recently, there has been a lot of debate over the validity of F-ST (and its analogue G(ST)) as a measure for population genetic differentiation, notably since the proposal of the new index D in 2008. Although several papers reviewed or explored specific features of these statistical measures, a succinct account of this bewildering issue with an overall update appears to be desirable. This is the purpose of the present review. The available statistics generally fall into two categories, represented by F-ST and D, respectively. None of them is perfect in measuring population genetic differentiation. Nevertheless, they each have advantages and are valuable for current research. In practice, both indices should be calculated and a comparison of them can generate useful insights into the evolutionary processes that influence population differentiation. F-ST (G(ST)) has some unique irreplaceable characteristics assuring its standing as the default measure for the foreseeable near future. Also, it will continue to serve as the standard for any alternative measures to contrast with. Instead of being anxious about making choice between these indices, one should pay due attention to the equilibrium status and the level of diversity (especially H-S) of the populations, since they largely sway the power of a given statistic to address a specific question. We provide a multi-faceted comparative summary of the various statistics, which can serve as a basic reference for readers to guide their applications [Current Zoology 61 (5): 886-897, 2015].