Abstract

Uncovering the genetic basis of adaptation is one of the greatest challenges facing modern evolutionary biology, often sparking much controversy, especially among practitioners within different subfields. Here we ask where is the locus of adaptation from the perspective of ecological genomics (ecogen) and evolutionary developmental biology (evodevo). Ecogen focuses on identifying loci under selection between populations living in different environments by scanning genome-wide patterns of genetic divergence, while evodevo focuses on candidate developmental regulatory genes and networks underlying phenotypic differences between species and higher taxa. We attempt to reconcile these alternative perspectives by studying the response of ant populations to past climate change on the Arizona Sky Islands, which are high elevation mountain ranges that represents a replicated natural experiment. We previously showed that adaptation to climatic changes in the Arizona Sky Islands in ant species Monomorium emersoni occurred through repeatable changes within the gene network underlying the development of dispersal alternative phenotypes: winged and wingless queens. Here, we uncover several loci under positive selection which associate with habitat temperature, a pattern consistent with a predictable and repeated increase in frequency following climatic changes on each of the Sky Islands. Furthermore, we show that gene flow between locations within a Sky Island is not a consequence of dispersal phenotypes and its gene network, but rather, is restricted by temperature adaptation along the ecological gradient. This suggests that the determination of winged and wingless queens may be developmentally plastic, and this plasticity and may facilitate jumps between adaptive peaks on the fitness landscape. This complex interplay promotes predictability on both genetic and phenotypic levels for multiple traits in the same species.