The population genomics of sexually antagonistic variation in Drosophila

Males and females of many species differ significantly in phenotype, sometimes so much so that the sexes were initially described as different species. Such sexual dimorphism is rooted in the differences between male and female reproductive roles, which select for different optimal morphologies, physiologies and behaviours in each sex. But while ubiquitous, sexual dimorphism is also often incomplete. Many populations harbour genetic variation that is 'sexually antagonistic', where alleles are beneficial to one sex but detrimental to the other. Sexual antagonism arises from the conflict of, on the one hand, genetic coupling between two sexes that share a genome and, on the other hand, opposing selection in the two sexes on homologous traits. As a result selection can maintain sexually antagonistic genetic variants in the population for prolonged periods of time. These variants are highly relevant because they are functionally important (after all, they affect fitness), are common in the population and benefit one sex while harming the other.

Far from an obscure by-product of sex-specific selection, sexual antagonism is an important evolutionary force. Due to their deleterious effects in one sex, sexual antagonistic variants result in partial maladaptation, with neither of the two sexes attaining its optimal phenotype. Sexual antagonism thus plays an important role in the maintenance of fitness variation, including sex-specific human disease alleles. SA is also a major driver in the evolution of differentiated sex chromosomes, further contributing to sex differences. And finally, SA is as model for adaptive conflicts in general, where the fitness associated with a genetic variant differs in different contexts. This includes, for example, alleles that are beneficial early in life and deleterious later on, or functional trade-offs in enzymes that are involved in multiple reactions. The study of SA thus has implications that go from sex differences and genome structure to human ageing and health.

Yet, despite its fundamental role as an evolutionary driver, the genetic bases of SA and its evolutionary dynamics remain poorly characterised. This is partly because historically we had limited information about the genetic basis of sexual antagonism. Only recently were we able to start solving this problem by identifying hundreds of sexually antagonistic variants in a laboratory population of fruit flies. This was a significant step forward and has opened the possibility to address the many gaps in our understanding of sexual antagonism and the limits to sex-specific adaptation. We can now ask questions about the fitness effects of individual antagonistic alleles in wild populations, gain insights into the evolutionary dynamics and turn-over of antagonistic variants, and explore the role that sexual antagonism plays in maintaining genetic variation within natural populations. We will answer these questions here, applying sophisticated computational population genomics approaches to hundreds of Drosophila genome sequences from around the world.

This work will represent a leap forward in our understanding of the genetics and evolution of sexual antagonism, and of the genetic processes that promote and limit sex-specific adaptation and the evolution of sexual dimorphism. Addressing these questions matters well beyond the field of evolutionary genetics, and our results will be relevant to animal breeders aiming to improve sex-specific traits, and biomedical researchers interested in the maintenance of risk factors for sex-specific disease.

Sexual antagonism (SA) arises when divergent selection pressures on males and females act on a genome that is shared between the sexes. This can favour the maintenance of antagonistic alleles, which have opposing fitness effects in the two sexes and can be maintained through balancing SA selection. SA limits genetic adaptation and maintains genetic variation for fitness and health. Despite these important effects, we know little about the genetic basis and evolutionary dynamics of SA. We recently took an important step to fill this gap of knowledge by identifying hundreds of SA polymorphisms in laboratory Drosophila. We are now in position to investigate how these SA alleles behave in natural populations: infer the fitness effects of SA alleles in the wild, investigate the evolutionary dynamics and turn-over of SA polymorphisms, and explore the role that SA plays in maintaining genetic variation in nature.

Here, we will address these questions by applying sophisticated computational population genomics approaches to hundreds of Drosophila genome sequences from around the world. Specifically, we will 1) apply Approximate Bayesian Computation (ABC) to infer the evolutionary parameters (sex-specific selection and dominance coefficients) of SA candidates across populations of fruit flies, to gain a detailed understanding of the properties of SA polymorphisms, 2) compare ABC estimates of the allelic ages of SA polymorphisms to their expected ages, to gain information about the lifespan and potential resolution of SA variants, and 3) apply a range of existing methods to identify targets of balancing selection and gain insights into the role of SA a force that maintains polymorphism in the genome.

This work will represent a significant step forward in our understanding of the genetics and evolution of SA fitness variation, and hence of the fundamental genetic constraints that limit sex-specific adaptation and the evolution of sexual dimorphism.