Sex-biased admixture and assortative mating shape genetic variation and influence demographic inference in admixed Cabo Verdeans

Abstract

Genetic data can provide insights into population history, but first, we must understand the patterns that complex histories leave in genomes. Here, we consider the admixed human population of Cabo Verde to understand the patterns of genetic variation left by social and demographic processes. First settled in the late 1400s, Cabo Verdeans are admixed descendants of Portuguese colonizers and enslaved West African people. We consider Cabo Verde's well-studied historical record alongside genome-wide SNP data from 563 individuals from 4 regions within the archipelago. We use genetic ancestry to test for patterns of nonrandom mating and sex-specific gene flow, and we examine the consequences of these processes for common demographic inference methods and genetic patterns. Notably, multiple population genetic tools that assume random mating underestimate the timing of admixture, but incorporating nonrandom mating produces estimates more consistent with historical records. We consider how admixture interrupts common summaries of genomic variation such as runs of homozygosity. While summaries of runs of homozygosity may be difficult to interpret in admixed populations, differentiating runs of homozygosity by length class shows that runs of homozygosity reflect historical differences between the islands in their contributions from the source populations and postadmixture population dynamics. Finally, we find higher African ancestry on the X chromosome than on the autosomes, consistent with an excess of European males and African females contributing to the gene pool. Considering these genomic insights into population history in the context of Cabo Verde's historical record, we can identify how assumptions in genetic models impact inference of population history more broadly.

Keywords: admixture; assortative mating; demographic inference; runs of homozygosity; sex bias.

Fig. 1.

Shared ancestry between and within the islands in the context of geography. Each island has a corresponding cluster of nodes representing all sampled individuals, with the individuals localized to be adjacent to the island where they were sampled. Node placement within islands is determined by a force-directed algorithm using pairwise shared IBD, meaning that the spread of each cluster reflects the level of relatedness in each population. Edges between the nodes represent the total IBD tract length for pairs of individuals sharing more than 150 cM of total IBD, with edges colored using the log-transformed total IBD length.

Fig. 2.

Correlation between inferred parental ancestries. a) For each Cabo Verdean individual, the inferred Parent 1 vs Parent 2 ancestry proportions are shown for an example chromosome (Chromosome 7), colored by island. b) The set of correlation coefficients between inferred parental ancestries for each autosomal chromosome.

Fig. 3.

Timelines of inferred generations of admixture for each island of Cabo Verde. The island regions are listed in order of known settlement timing beginning with Santiago (a), followed by Fogo (b), and then the later migrations to the Northwest Cluster (c) and Boa Vista (d). For each island, admixture timing inferred with population genetic methods are shown in comparison to historical records. Historical estimates of when each island were first settled are shown in generations, based on a range of generation times (20–30 year generation time). For the LAD-based estimates of admixture timing, we tested a range of ancestry-assortative mating strengths. Specifically, we tested an assortative mating strength of 0 (results shown in green) vs values inferred using ANCESTOR in Fig 2 (results shown in blue). Within each of these LAD-based estimates of admixture timing, the presented interval reflects the range of estimates generated under the assumption of constant migration rate (m = 0.01; yielding older estimates) to the assumption of no migration (m = 0; yielding more recent estimates). MultiWaver results show the admixture generations inferred under the admixture model selected by MultiWaver, with 0.95 CI.

Fig. 4.

ROH content by island and in the context of ancestry. The total length of ROH per individual genome is plotted under. (a) GARLIC-inferred ROH class boundaries (shorter/long ROH boundary of approximately 1.08 Mb for Santiago, 1.08 Mb for NW Cluster, 1.12 Mb for Fogo, 1.14 Mb for Boa Vista, 1.26 Mb for West African reference individuals, and 0.91 Mb for European reference individuals); and (b) under class boundaries estimated based on the number of admixture generations (shorter/long ROH boundary of 2.5 Mb). Violin plots show the population-specific distributions of the total (summed over each genome) length of autosomal ROH per individual. Solid black dots represent the within-population means. Scatter plots show the total length of autosomal ROH per individual plotted against West African ancestry proportions and colored by population.

Fig. 5.

Sex-biased admixture in Cabo Verde. (a) The distribution of West African ancestry proportion on the autosomes and X chromosome for each of the island regions, estimated with ADMIXTURE. (b) Under a model of constant admixture over time, the fraction of the total contribution of genetic material originating from females for West African and European source populations. Here, we show the distribution of parameter sets for the smallest 0.1% of Euclidean distances between the model-predicted and observed X and autosomal ancestry from a grid of possible parameter values. The range of sex-specific contributions from West African and European source populations that produce ancestry estimates closest to those observed in Cabo Verde is shown for Cabo Verde as a whole (left), and then broken down by region, with medians (dashed lines).

Fig. 6.

Autosomal (chromosome 7) vs X chromosome distributions of ROH by class. a) The population-specific distributions of the total length of shorter ROH per individual on the X chromosome compared to an autosome. Chromosome 7 was chosen as the autosomal point of comparison, given that it is the autosome most similar in size to the X chromosome. Totals are shown for females only, so that the same samples are being compared across the X chromosome and chromosome 7. Totals are plotted per Mb to account for slight differences in chromosome lengths. b) The same information for long ROH.