Recessive deleterious variation has a limited impact on signals of adaptive introgression in human populations

Abstract

Admixture with archaic hominins has altered the landscape of genomic variation in modern human populations. Several gene regions have been previously identified as candidates of adaptive introgression (AI) that facilitated human adaptation to specific environments. However, simulation-based studies have suggested that population genetics processes other than adaptive mutations, such as heterosis from recessive deleterious variants private to populations before admixture, can also lead to patterns in genomic data that resemble adaptive introgression. The extent to which the presence of deleterious variants affect the false-positive rate and the power of current methods to detect AI has not been fully assessed. Here, we used extensive simulations to show that recessive deleterious mutations can increase the false positive rates of tests for AI compared to models without deleterious variants. We further examined candidates of AI in modern humans identified from previous studies and show that, although deleterious variants may hinder the performance of AI detection in modern humans, most signals remained robust when deleterious variants are included in the null model. While deleterious variants may have a limited impact on detecting signals of adaptive introgression in humans, we found that at least two AI candidate genes, HYAL2 and HLA, are particularly susceptible to high false positive rates due to the recessive deleterious mutations. By quantifying parameters that affect heterosis, we show that the high false positives are largely attributed to the high exon densities together with low recombination rates in the genomic regions, which can further be exaggerated by the population growth in recent human evolution. Although the combination of such parameters is rare in the human genome, caution is still warranted in other species with different genomic composition and demographic histories.