. 2020 Jul 6;20(6):1517–1525. doi: 10.1111/1755-0998.13210

TABLE 2.

Software packages for estimating F _ST and their estimates using mock input. These input files contained fixed differences between the X‐ and Y‐chromosome for various sample sizes of males and females

Package (version)	Option	5 Males and 5 females	10 Males and 10 females	20 Males and 20 females	5 Males and 15 females	15 Males and 5 females	Referenced estimator
vcftools (0.1.15)	weir‐fst‐pop	0.5	0.5	0.5	0.667	0.4	Weir and Cockerham, (1984)
popgenome (2.61)	F_ST.stats: nucleotide.F_ST	0.444	0.474	0.487	0.444	0.483	Hudson, Slatkin, et al. (1992)
	F_ST.stats: nuc.F_ST.pairwise	0.444	0.474	0.487	0.444	0.483	Hudson, Slatkin, et al. (1992)
	F_ST.stats: Nei.G_ST	0.333	0.333	0.333	0.333	0.333	Nei (1973)
	F_ST.stats: Nei.G_ST.pairwise	0.333	0.333	0.333	0.333	0.333	Nei (1973)
	F_ST.stats: Hudson.H_ST	0.296	0.316	0.325	0.45	0.163	Hudson, Boos, et al. (1992) ^a
	F_ST.stats: Hudson.G_ST	0.286	0.310	0.322	0.378	0.195	Hudson, Boos, et al. (1992) ^b ^, ^c
diversity (1.9.90)	diffCalc(fst = TRUE)	0.5	0.5	0.5	0.667	0.4	Weir and Cockerham (1984)
diversity (1.9.90)	diffCalc()	0.286	0.310	0.322	0.3023	0.3023	Nei and Chesser (1983)
hierfstat (0.04–29)	pairwise.fst	0.333	0.333	0.333	0.429	0.2	Nei (1973) ^c
	genet.dist(method = Nei87)	0.5	0.5	0.5	0.5	0.5	Nei (1987) ^d
	pairwise.neifst	0.5	0.5	0.5	0.5	0.5	Nei (1987) ^d
	basic.stats(fst)	0.333	0.333	0.333	0.333	0.333	Nei (1987)
	genet.dist(method = WC84)	0.5	0.5	0.5	0.667	0.4	Weir and Cockerham, (1984)
	pairwise.WCfst	0.5	0.5	0.5	0.667	0.4	Weir and Cockerham, (1984)
genepop (1.0.5)	Fst	0.5	0.5	0.5	0.667	0.4	Weir and Cockerham, (1984)
arlequin (3.5)	Compute pairwise FST	0.444	0.474	0.487	0.647	0.362	Excoffier, Smouse, and Quattro (1992)
dnasp (6.12.03)	Gene Flow and Genetic Differentiation: GST	0.286	0.310	0.322	0.378	0.194	Nei (1973) ^b ^, ^c
	Gene Flow and Genetic Differentiation: GammaSt	0.333	0.333	0.333	0.429	0.2	Nei (1982) ^c
	Gene Flow and Genetic Differentiation: Fst	0.444	0.474	0.487	0.444	0.483	Hudson, Slatkin, et al. (1992)

^{^a}

This implementation appears to use a $w_{i} = \frac{n_{i} ‐ 2}{n_{1} + n_{2} ‐ 4}$ weighting factor.

^{^b}

These estimates are most consistent with Nei and Chesser (1983), which is also discussed in Hudson, Boos, et al. (1992).

^{^c}

These metrics appear to use a $w_{i} = \frac{n_{i}}{n_{1} + n_{2}}$ weighting factor, while Nei (1982) and Nei and Chesser (1983) state that in most practices the subpopulations can be assumed to be weighted equally.

^{^d}

The referenced estimator is consistent with F′_ST in Nei (1987).