The fitness of an introgressing haplotype

The genomic era has made clear that introgression, or the movement of genetic material between species, is a common feature of evolution. Examples of both adaptive and deleterious introgression exist in a variety of systems. What is unclear is how the fitness of an introgressing haplotype changes as species diverge, or as the size of the introgressing haplotype changes. In a simple model, we show that early in the process of divergence, introgression of large haplotypes can be favored more than introgression of individual alleles. The key insight is that alleles from a shared genetic background are likely to have positive epistatic interactions, increasing the fitness of a larger introgressing block. The buildup of incompatibilities between diverging species in the form of deleterious epistasis eventually favors the introgression of small haplotypes as the number of diverged alleles increases, and eventually even single alleles with positive direct effects can be selected against. This model is consistent with observations of a positive relationship between recombination rate and introgression frequency across the genome, however it generates several novel predictions. First, the model suggests that the relationship between recombination rate and introgression may not exist, or may be negative, in recently diverged species pairs. Furthermore, the model suggests that introgression that replaces existing derived variation will always be more deleterious than introgression at sites carrying ancestral variants. These predictions are tested in an example of introgression in D. melanogaster, with some support for both.