Figure 1.
Geographic location of the samples analyzed in this study.
141 samples for the Frequency dataset (A) and 42 samples for the Genotype dataset (B). The populations are assigned to 11 and 8 major geographic groups, respectively.
Figure 2.
Average genetic diversity (and standard deviation) over populations in geographic groups.
10 groups for the Frequency dataset and 8 for the Genotype dataset (see text). A and C graphs show the distribution of the variance in allele sizes (Vp) for the Frequency and Genotype datasets, respectively. B and D graphs show the distribution of the expected heterozygosity (He) for the Frequency and Genotype datasets, respectively. P-values for the Kruskal-Wallis test (test of significant differences among all groups).
Figure 3.
Plots of population diversity against geographic distance to East Africa (Addis Ababa).
A: Vp against geographic distance for the Frequency dataset; B: He against geographic distance for the Frequency dataset; C: Vp against geographic distance for the Genotype dataset; D: He against geographic distance for the Genotype dataset. The determination coefficient (R2) estimates the proportion of the variation in genetic diversity that is explained by the variation in geographic distance to East Africa.
Figure 4.
Plots of the multidimesional (MDS) scaling analyses of genetic distances inferred from the forensics datasets.
A: MDS of genetic distances computed on the Frequency dataset (stress = 0.18); B: MDS of genetic distances computed on the Genotype dataset (stress = 0.13). Shown in caption: population samples are color-coded following the 12 main geographic groups listed in Figure 1.
Table 1.
Comparison of AMOVA results across studies.
Figure 5.
Distributions of the mean number of alleles and with He for various datasets.
A. Distribution over loci of the mean number of alleles per sample in the two forensics tetra STRs datasets (Frequency and Genotype) and in the HGDP dataset and subsets of tetra STRs published by Pemberton et al. [15] (complete tetra STRs dataset of 434 loci, and subsets of 13 loci biased towards high or low He, see text). B. Distribution of He over the number of alleles for each locus in each sample.