Human Biology Open Access Pre-Prints

Document Type

Open Access Preprint

Anticipated Volume


Anticipated Issue



Objectives. Geneticists have argued that the linear decay in within-population genetic diversity with increasing geographic distance from East Africa is best explained by a phylogenetic process of founder effects, growth, and isolation termed serial founder effects (SFE). However, the SFE process has not yet been adequately vetted against other evolutionary processes that may also affect geospatial patterns of diversity. Additionally, studies of SFE have been largely based on a limited 52 population sample from the HGDP-CEPH. Here, we assess the effects of SFE, admixture, and localized gene flow processes on patterns of global and regional diversity using a published dataset consisting of 645 autosomal microsatellite genotypes from 5,415 individuals in 248 widespread populations.

Materials and Methods. Because SFE is a phylogenetic process, we used a formal tree-fitting approach to explore the role of the process in shaping patterns of global and regional diversity. The approach involved fitting global and regional population trees to extant patterns of gene diversity and then systematically examining the deviations in fit. We also informally tested the SFE process using linear models of gene diversity vs. waypoint geographic distances from Africa. Because gene flow and phylogenetic processes can both shape geospatial patterns of diversity, we tested the role of localized gene flow using partial Mantel correlograms of gene diversity vs. geographic distance controlling for the confounding effects of tree-like genetic structure.

Results. We corroborate previous findings that global patterns of diversity, both within and between populations, are the product of an out-of-Africa SFE process. Within regions, however, diversity within populations is uncorrelated with geographic distance from Africa. Instead, deviations in the fit of regional population trees are largely the product of recent inter-regional admixture. Additionally, in several regions, we found that positive correlations between pairwise gene diversity and geographic distance, frequently attributed to localized gene flow, were instead the product of phylogenetic processes associated with initial peopling or subsequent range expansions.

Conclusions. Detailed analyses of the pattern of diversity within and between populations reveal that the signatures of different evolutionary processes dominate at different geographic scales. These findings have important implications for recent publications on the biology of race.