Table 3.
Analysis of molecular variance (AMOVA) under four criteria of classification of populations, based on the 870 bp long NAT2 coding exon
| Percentage of variation | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Dataset1 | Categories grouping2 | Number of population samples | Number of groups | Between groups | Between populations within groups | Fixation indexes | |||
| ΦCT | P-value3 | ΦSC | P-value4 | ||||||
| AFR | Geo5 | 37 | 3 | −0.016 | 3.23 | −0.000126 | 0.3722 | 0.03230 | <0.0001 (0.0004) |
| Lang | 38 | 4 | 0.43 | 3.04 | 0.00425 | 0.1323 | 0.03053 | <0.0001 (0.0004) | |
| Subsist | 38 | 4 | 1.82 | 2.02 | 0.01819 | 0.0003 (0.0012) | 0.02055 | <0.0001 (0.0004) | |
| Clim | 38 | 2 | 2.35 | 2.11 | 0.02345 | 0.0001 (0.0004) | 0.02161 | <0.0001 (0.0004) | |
| FP | Geo5 | 28 | 3 | −0.046 | 2.21 | −0.000376 | 0.4572 | 0.02207 | <0.0001 (0.0004) |
| Lang | 29 | 3 | −0.216 | 2.30 | −0.002076 | 0.6201 | 0.02297 | <0.0001 (0.0004) | |
| Subsist | 29 | 3 | 1.16 | 1.58 | 0.01156 | 0.0251 (0.1004) | 0.01600 | 0.0001 (0.0004) | |
| Clim | 29 | 2 | 3.57 | 0.71 | 0.03572 | <0.0001 (0.0004) | 0.00739 | 0.0170 (0.0680) | |
| FPLS | Geo | 13 | 3 | −0.536 | 2.99 | −0.005276 | 0.7238 | 0.02974 | <0.0001 (0.0004) |
| Lang7 | 12 | 2 | −0.456 | 2.84 | −0.004476 | 0.5612 | 0.02832 | <0.0001 (0.0004) | |
| Subsist | 13 | 2 | 1.20 | 1.99 | 0.01202 | 0.0530 | 0.02016 | <0.0001 (0.0004) | |
| Clim | 13 | 2 | 5.40 | 0.54 | 0.05403 | 0.0035 (0.0140) | 0.00568 | 0.0462 (0.1848) | |
1 As described in Methods, the three population data subsets are AFR: 38 samples, excluding the Americans of African ancestry (ASW of [53], see Table 1); FP: 29 samples of African food-producing populations; FPLS: 13 samples of African food-producing populations with sample size ≥ 20 individuals. Thus, the ASW sample was not considered in any of the AMOVA analyses
2 Categories as in Table 1 : classification according to geographical region (Geo), subsistence mode (Subsist), linguistic affiliation (Lang), and ecoregion (Clim), namely climatic zone and biome, which defines the fourth categorization criterion that considers whether populations live within the dry savanna biome or outside of it
3 Significance of the ΦCT index and of the corresponding percentage of variation due to differences between groups. Significant P-values (i.e., <5 %) are shown in bold, and adjusted P-values after Bonferroni correction for multiple testing (here, four tests) are provided in brackets
4 Significance of the ΦSC index and of the corresponding percentage of variation due to differences between populations within groups. Significant P-values (i.e., <5%) are shown in bold, and adjusted P-values after Bonferroni correction for multiple testing (here, four tests) are provided in brackets
5 Only three geographical regions are considered here (Western, Central and Eastern, Table 1) because the fourth region (Northern) is represented by one population sample only (EGY)
6 Because variance components in AMOVA are actually defined as covariances, negative values can occur [95]. A negative ΦCT value would be expected if gene copies were more correlated between groups than between populations within groups. However, none of the negative ΦCT values in the table are statistically significant, thus indicating that they are equal to zero
7 Only two linguistic families are considered here (Niger-Congo and Nilo-Saharan, Table 1) because the third family (Afro-Asiatic) is represented by one population sample only (SOM)