The influence of clan structure on the genetic variation in a single Ghanaian village

Abstract

Socioeconomic and cultural factors are thought to have an important role in influencing human population genetic structure. To explain such population structure differences, most studies analyse genetic differences among widely dispersed human populations. In contrast, we have studied the genetic structure of an ethnic group occupying a single village in north-eastern Ghana. We found a markedly skewed male population substructure because of an almost complete lack of male gene flow among Bimoba clans in this village. We also observed a deep male substructure within one of the clans in this village. Among all males, we observed only three Y-single-nucleotide polymorphism (SNP) haplogroups: E1b1a*-M2, E1b1a7a*-U174 and E1b1a8a*-U209, P277, P278. In contrast to the marked Y-chromosomal substructure, mitochondrial DNA HVS-1 sequence variation and autosomal short-tandem repeats variation patterns indicate high genetic diversities and a virtually random female-mediated gene flow among clans. On the extreme micro-geographical scale of this single Bimoba village, correspondence between the Y-chromosome lineages and clan membership could be due to the combined effects of the strict patrilocal and patrilineal structure. If translated to larger geographic scales, our results would imply that the extent of variation in uniparentally inherited genetic markers, which are typically associated with historical migration on a continental scale, could equally likely be the result of many small and different cumulative effects of social factors such as clan membership that act at a local scale. Such local scale effects should therefore be considered in genetic studies, especially those that use uniparental markers, before making inferences about human history at large.

Figure 1

The geographical distribution of compounds (circles) in the village. Compounds are colour-labelled according to: (a) the Y-sub-haplogroup within Y-E1b1a*-M2, (b) the reported clan. The lines correspond to the spatial classification of the compound labels with a density-based Parzen classifier. The diameter of each circle is proportional to the number of sampled males in that compound; the smallest diameter represents one male, the largest represents nine males. The colour coding in (b) corresponds to that used in Figure 2.

Figure 2

Median joining networks for (a) Y-STR haplotypes, and (b) mtDNA HVS1 sequence haplotypes. Each haplotype pie is colour-labelled according to the reported clan of males carrying that haplotype; the colour labels correspond to those in Figure 1b. In the Y-STR haplotype network (a), the segments with the same Y-haplogroup are indicated by a similar background colour that corresponds to the colour labels in Figure 1a. The smallest distance between two haplotypes equals one repeat length difference in one STR. The diameter of each pie is proportional to the frequency of that haplotype; the smallest pies represent a single male, the largest 81 males.