Out of a table comprising allelic data of at least one locus for several
individuals and from at least two populations, measurements of genetic
differentiation (i.e. D, Dest, Dest.Chao, Gst and Gst.est (Jost, 2008))
can be calculated. P-values and 95% confidence intervals are calculated
according to Manly (1997) using a bootstrap method. Thereby, for each
locus, alleles are reallocated over populations if they are all in Hardy
Weinberg Equilibrium (HWE). If not, the genotypes are reallocated, since
alleles can not be assumed to be inherited independently from each other
(Goudet, 1996). The p-values state how much the empirical found allocation
of alleles over populations differs from an allocation that could arise by
chance alone. In the case that more than two populations are compared
pairwise, the p-values are adjusted besides using bonferroni correction in
several other ways due to the multiple comparison from one data
set.