Adaptive genomic variation associated with environmental gradients along a latitudinal cline in Rana temporaria

Abstract

Rana temporaria occur across a large geographic and environmental gradient in Scandinavia. Several studies involving common garden experiments have established adaptive divergence across the gradient. The main objective of this study was to determine the extent of neutral and adaptive genetic divergence across the latitudinal gradient. Here we sequence genome-wide markers for 15 populations from six regions sampled from southern Sweden to Finland. Using a multivariate approach we find that 68% of the genomic variation is associated with climate or geographically structured climate. Using outlier scans and environmental association analyses we identify a set of potentially adaptive loci and examine their change in allele frequency associated with different climatic variables. Using a gradient forest analysis we identify points along three of the climate variables where allele frequencies change more rapidly than expected if it were a linear association. We identify a large threshold effect associated with BIO5 (mean temperature during the warmest month) which is seen as a rapid change in southern Sweden. By comparing the change in neutral and adaptive allele frequencies across the whole gradient, we identify southern Sweden as a region with the largest divergence between the datasets. This suggests small changes in the climate may result in a mismatch between the adaptive genotypes and the environment in these populations. Overall this study shows that genomic analyses can provide a powerful complement to common garden experiments to improve our understanding of adaptive divergence across heterogeneous landscapes.