This repository contains the simulation, plotting, and tex code used in the production of A few stickleback can transport adaptive alleles to new lakes
This paper had been accepted by the journal Genes, Genomes, Genetics The Arxiv of this paper can be found here
Threespine stickleback provide a striking example of local adaptation to divergent habitats while still being connected by recurrent gene flow. These small fish occur in marine and freshwater habitats throughout the Northern Hemisphere, and in numerous cases smaller freshwater populations have been established ``de novo'' from marine colonists. Previous results show that a similar freshwater phenotype often evolves in independent populations soon after colonization using largely the same standing genetic variation, even in very distant geographic regions. But how many stickleback migrants are needed to capture a sufficient amount of standing variation to fuel adaptation? Here we use genealogical simulations to determine the levels of gene flow that best match observed patterns of allele sharing among habitats in stickleback, and to better understand how gene flow and local adaptation in large metapopulations determine the speed of adaptation and re-use of standing genetic variation. In our simulations we find that repeated adaptation uses a shared set of alleles that are maintained at low frequency by migration-selection balance in ocean populations. This process occurs over a realistic range of intermediate levels of gene flow that match previous empirical population genomic studies in stickleback. Examining these simulations more deeply reveals that low gene flow leads to slow, independent adaption to district habitats, whereas high gene flow leads to significant mutation load but an increased probability of successful population genomic scans for locally adapted alleles. We also find that the original generation of marine individuals that colonized the lakes is most often the sole source of most freshwater adapted alleles. These simulations provide deeper context for existing studies of stickleback evolutionary genomics, and provide guidance for future empirical studies in this model. More broadly, our results support existing theory of local adaptation but extend it by more completely documenting the genealogical history of adaptive alleles in a metapopulation.