Poster Presentation Society for Molecular Biology and Evolution Conference 2016

Beyond clines: lineages and haplotype blocks in hybrid zones (#652)

Alisa Sedghifar 1 , Yaniv Brandvain 2 , Peter Ralph 3
  1. Lewis-Sigler Institute for integrative genomics, Princeton University, Princeton, NJ, United States
  2. Plant Biology, University of Minnesota, St Paul, MN, United States
  3. Computational Biology and Bioinformatics, University of Southern California, Los Angeles, CA, United States

Hybrid zones formed between recently diverged populations offer an opportunity to study the mechanisms underlying reproductive isolation and the process of speciation. In particular, selection against hybrids can impact genome-wide patterns of introgression. Loci associated with incompatibilities are rapidly removed by selection when in the wrong species background and as a result, in populations away from the hybrid zone center, will be present in lineages that are relatively recent migrants. We expect this to be reflected in longer tracts of ancestry surrounding targets of selection.

To test this prediction, we use a combination of analytical theory and simulations to describe the movement and breaking up of lineages as a result of migration, hybridization and recombination. Under our Brownian model, we find that blocks of ancestry surrounding single-locus incompatibilities can be substantially longer than the genome-wide average, and that locally disadvantageous alleles tend to exist as smaller families.

These patterns may be used to characterize the age of hybrid zones and to identify targets of selection, thereby deepening our understanding of the genetic factors driving lineage-specific adaptation and reproductive isolation between species. Studies of selection in hybrid zones have traditionally focused on cline width at individual loci, and the availability of population genomic data now enables use of the additional information contained in tracts of unbroken ancestry to better understand the processes operating in these populations.