W606
Parallel Genomic Evolution But Not Parallel Speciation in Annual Sunflowers

Date: Saturday, January 12, 2013
Time: 3:55 PM
Room: Pacific Salon 3
Sebastien Renaut , University of British Columbia, Vancouver, BC, Canada
Loren Rieseberg , University of British Columbia, Vancouver, BC, Canada
The late Stephen Gould argued that “rewinding the tape of life” would produce a very different array of end products, but others have reported that the evolution of some traits and species can be surprisingly repeatable. However, little is known about the repeatability of evolution at the genomic level. Here we employ transcriptome sequences from two pairs of sister species in the sunflower genus (Helianthus annuus - H. argophyllus and H. debilis - H. petiolaris) to study the repeatability of genomic changes that have accompanied their divergence.  Analyses of divergence indicated high repeatability across 1 cM genomic windows (r2 = 0.33), modest repeatability at the level of individual genes (r2 = 0.14) and no repeatability at the SNP level. However, when analyzing only the small fraction of outlier SNPs, we find that these are shared between the diverging lineages circa 50% more than expected by chance.  So why are patterns of genomic divergence repeatable?  While the species’ pairs targeted by this study do not represent products of parallel speciation events, they have diverged along a similar photoperiod, temperature, and moisture gradient.  Thus, parallel selection pressure probably accounts for the parallel divergence of outlier SNPs.  Also, we suspect that some genes matter more than others for adaptation and that such importance is correlated across species.  Lastly, independently diverging lineages likely share heterogeneity in recombination rates, which in turn, strongly influences patterns of divergence among sunflower species pairs.  Thus, parallel genomic evolution results from both parallel adaptation and shared genomic architecture.