To Migrate or not to Migrate? Exploring the Genomic Basis of Partial Migratory Behavior in Bats.

Migration is a widespread phenomenon in animals that involves the synchronized movement of numerous individuals across habitats. While migratory traits appear to be environmentally triggered, evidence also points to a still poorly understood genetically regulated mechanism. The study of both the genomic architecture of migration and the degree of similarity across migrating taxa is a recurrent topic in evolutionary biology. Here, we investigated the genomic basis of migration in a flying mammal, the lesser long-nosed bat (Leptonycteris yerbabuenae), a nectar-feeding bat with a partially migratory behavior. Each year, the migrant group of females travels north from central Mexico to give birth in the Sonoran Desert, while the resident females remain and give birth in central Mexico. Using RAD-seq, we detected a demographic decline in this species during the Last Glacial Maximum and found that resident and migratory females form a single genetic cluster. Nevertheless, we identified 10 divergent genomic regions enriched with highly differentiated SNPs (FST values three or more orders of magnitude above the mean). Seven of such regions bear signatures of balancing selection and contain genes that have been identified in other migrating animals; such genes are qualitatively enriched for nervous system-related functions, potentially linked to circadian cycle, orientation, and navigation. Given that the migratory behavior is supposed to have originated recently in L. yerbabuenae (i.e., 10 to 20 Ka), we hypothesize that strong diversifying selection is operating in specific regions of the genome, while the rest is homogenized by the effect of males, which mate indistinctively with females from both groups.