Does the paradigm of genotype-environment associations need to be re-assessed? The paradox of adaptive phenotypic clines with non-clinal patterns in causal alleles

Abstract

Multivariate climate change presents an urgent need to understand how species adapt to complex environments. Population genetic theory predicts that loci under selection will form monotonic allele frequency clines with their selective environment, which has led to the wide use of genotype-environment associations (GEAs). This study elucidates the conditions under which allele frequency clines are more or less likely to evolve as multiple quantitative traits adapt to a multivariate environment. A novel set of simulations was created that all evolved similar phenotypic clines, but with varying proportions of causal alleles with clines. Phenotypic clines evolved mostly without clines in the causal allele frequencies under conditions that promoted unique combinations of mutations to achieve the multivariate optimum in different parts of the landscape. Although univariate and multivariate GEA methods failed to accurately infer the genetic basis of adaptation under a range of scenarios, individual multivariate traits could be accurately predicted from genotype and environmental data without any knowledge of the genetic architecture. This research challenges the utility of GEAs for understanding the genetic basis of adaptation to the environment, and instead suggests that multivariate trait predictions are a more fruitful approach for genomic forecasting and assisted gene flow efforts.