Ancestry and Ecology but not Geography Explain Genome Correlations Indicative of Divergence in Mimulus aurantiacus

Abstract

One of the main goals of evolutionary biology is to understand the speciation process that creates the vast diversity of life we observe on earth. In the case of speciation with gene flow, it is important to understand how populations become isolated from each other to allow genetic differences to accumulate. This isolation can occur due to distance (IBD) and adaptation to the local ecology (IBA). While both have been characterized in different systems, it is unknown how these processes affect certain genome correlations involving FST that are indicative of linked selection and genome divergence. Using reduced representation sequencing (RAD-seq), we first characterized population structure in our study system Mimulus aurantiacus ssp. puniceus, where IBD and local adaptation have been well-characterized. We found that puniceus displays more levels of genetic subdivision than related subspecies, and we were able to characterize two distinct lineages of puniceus that display near-identical distributions of floral phenotypes. Our population genomics analysis then revealed that IBA was a better predictor of genome correlations than IBD in a subset of populations that occurred across a considerable climatic gradient. However, the best predictor of the genome correlations was differences in ancestry between the two discrete lineages characterized in our population structure analysis. These results show that ancestry differences can explain patterns of genomic divergence without clear signs of adaptation, but that divergent selection is likely important for explaining patterns of genomic divergence during early stages of speciation.