Skip to main content
Fig. 11 | BMC Bioinformatics

Fig. 11

From: Gene flow analysis method, the D-statistic, is robust in a wide parameter space

Fig. 11

Pairwise comparison between the f-statistics in the same scenario. Each box represents 500 × 500 = 250,000 data points. Upper-left triangle: probability of the higher real f value result in a statistically higher f-statistic, based on comparison of confidence intervals. Lower-right triangle: probability of the higher real f value result in a numerically higher f-statistic. Diagonal line (circles): probability of no statistically significant difference given the same real f values. Gray-shaded areas indicate real f values below the sensitivity of the D-statistic in the same scenario (mean of three replicates). In all graphs, the divergence time T3 = 1 × 105 generations, TGF = 0.5 T2 = 0.25 T3. a, b, c Ne = 0.2 T3; d, e, f Ne = T3; g, h, i Ne = 5 T3. a, d, g \( {\widehat{f}}_G \), gene flow direction H3 - > H2; b, e, h \( {\widehat{f}}_{hom} \), gene flow direction H3 - > H2; c, f, i \( {\widehat{f}}_{hom} \), gene flow direction H2 - > H3

Back to article page