History and structure of sub-Saharan populations of Drosophila melanogaster

Abstract

Drosophila melanogaster is an important model organism in evolutionary genetics, yet little is known about the population structure and the demographic history of this species within sub-Saharan Africa, which is thought to contain its ancestral range. We surveyed nucleotide variation at four 1-kb fragments in 240 individual lines representing 21 sub-Saharan and 4 Palearctic population samples of D. melanogaster. In agreement with recent studies, we find a small but significant level of genetic differentiation within sub-Saharan Africa. A clear geographic pattern is observed, with eastern and western African populations composing two genetically distinct groups. This pattern may have resulted from a relatively recent establishment of D. melanogaster in western Africa. Eastern populations show greater evidence for long-term stability, consistent with the hypothesis that eastern Africa contains the ancestral range of the species. Three sub-Saharan populations show evidence for cosmopolitan introgression. Apart from those cases, the closest relationships between Palearctic and sub-Saharan populations involve a sample from the rift zone (Uganda), suggesting that the progenitors of Palearctic D. melanogaster might have come from this region. Finally, we find a large excess of singleton polymorphisms in the full data set, which is best explained by a combination of population growth and purifying selection.

Figure 1.—

Geographic location of population samples used in this study. Population samples are identified by the two-letter abbreviations given in Table 1.

Figure 2.—

Unrooted population distance tree generated by neighbor-joining analysis of F_ST values. Distance is indicated by the lengths of horizontal branches. Putatively admixed populations (Er, Ga, SA) are excluded.

Figure 3.—

Population membership in the genetic clusters identified by Structure. In each case, one of the two clusters is arbitrarily shown as shaded, the other as open. (A) Population ancestry for the K = 2 Structure analysis of cosmopolitan and sub-Saharan samples. (B) Population ancestry for the K = 2 Structure analysis of nonadmixed sub-Saharan samples only.

Figure 4.—

The relationship between interspecific divergence and proportion of singleton polymorphisms for 100-bp sections of the loci studied. The dashed line indicates the proportion of singletons expected in a neutral equilibrium population.

Figure 5.—

Hypothesized history of D. melanogaster and related species. (A) EOTYMS refers to the common ancestor of D. erecta, D. orena, D. teissieri, D. yakuba, D. melanogaster, and D. simulans, (along with the three known island endemic species), which reached Africa <20 MYA. (B) The split between the ancestor of D. melanogaster and D. simulans (MS) and the ancestor of D. erecta, D. orena, D. teissieri, and D. yakuba (EOTY) occurred ∼10–15 MYA, possibly associated with emergence of the East African Rift. (C) The divergence of D. melanogaster and D. simulans 2–6 MYA may have been initiated by the dispersal of the latter species to Madagascar or other Indian Ocean islands. (D) More recently, D. melanogaster expanded from eastern Africa into western Africa and Eurasia, and D. simulans expanded into continental Africa and Eurasia.