Patterns of and processes shaping population structure and introgression among recently differentiated Drosophila melanogaster populations

Abstract

Understanding the factors that produce and maintain genetic variation is a central goal of evolutionary biology. Despite a century of genetic analysis, the evolutionary history underlying patterns of exceptional genetic and phenotypic variation in the model organism Drosophila melanogaster remains poorly understood. In particular, how genetic and phenotypic variation is partitioned across global D. melanogaster populations, and specifically in its putative ancestral range in Subtropical Africa, remains unresolved. Here, we integrate genomic and behavioral analyses to assess patterns of population genetic structure, admixture, mate preference, and genetic incompatibility throughout the range of this model organism. Our analysis includes 174 new accessions from novel and under-sampled regions within Subtropical Africa. We find that while almost all Out of Africa genomes correspond to a single genetic ancestry, different geographic regions within Africa contain multiple distinct ancestries, including the presence of substantial cryptic diversity within Subtropical Africa. We find evidence for significant admixture- and variation in admixture rates-between geographic regions within Africa, as well as between African and non-African lineages. By combining behavioral analysis with population genomics, we demonstrate that female mate choice is highly polymorphic, behavioral types are not monophyletic, and that genomic differences between behavioral types correspond to many regions across the genome. These include regions associated with neurological development, behavior, olfactory perception, and learning. Finally, we discovered that many individual pairs of putative incompatibility loci likely evolved during or after the expansion of D. melanogaster out of Africa. This work contributes to our understanding of the evolutionary history of a key model system, and provides insight into the distribution of reproductive barriers that are polymorphic within species.

Divergent selection on behavioural and chemical traits between reproductively isolated populations of Drosophila melanogaster

“…We identified large differences in male behaviours and CHCs among African strains, such that we could not define a single behavioural trait and CHC combination that classifies African vs non‐African males. This intriguing result suggests that there are segregating preferences in these populations, which could be a product of complex genetic structure within Southern Africa (Begun & Aquadro, 1993; Coughlan et al, 2021; Dieringer et al, 2005; Pool et al, 2012). This variation may have facilitated the rapid divergence of the non‐African lineages from African ancestors when these lineages migrated out of Africa.…”

“…melanogaster , we now understand that the relationship between Z‐type and M‐type lineages is more complicated. For example, there is not a single genetic lineage in southern Africa that corresponds to Z‐type, instead the behaviour of female rejection of non‐African males, is spread across populations and lineages (Coughlan et al, 2021). In contrast, out‐of‐Africa strains are more homogenous, likely representing a bottleneck from their migration event (Kapopoulou et al, 2018; Li & Stephan, 2006).…”

The neurodevelopmental genesalan shepardandNeurogliancontribute to female mate preference in AfricanDrosophila melanogaster

“…For Nrg there were two intronic SNPs that were in the top 1% and these SNPs were within 2 kb of each other. Interestingly these loci were also outliers in the study by Coughlan et al (2021) with shep in the top 1% of outliers and Nrg in the top 5% percentile. Previous behavioral analysis of homozygous viable mutant strains for these loci, found evidence that these genes were involved female behavior (Carhan et al, 2005; Chen et al, 2014).…”

“…The two male strains used in the experiment are phenotypically indistinguishable, so to be able to tell them apart we placed them on food containing blue food coloring 48 hours prior to the bheavioral assay. This is a non-invasive, robust identification method that does not affect male courtship behavior or female preference (Wu et al, 1995; Hollocher et al, 1997; Coughlan et al, 2021; Serrato-Capuchina et al, 2021; Jin et al, 2022). All behavioral assays were performed in a room held at a constant 20C degrees Celsius and 50% humidity within two hours of the incubator lights turning on to maximize the number of copulations we could observe.…”

“…Our goal was to identify outlier genes that showed a steeper cline in African ancestry compared to the genome wide average. We could also compare our list of candidates to independent analyses that used populations from Africa to identify female preference alleles (Coughlan et al, 2021).…”