Sex chromosomes and species barriers

Speciation is a gradual process that requires the interruption of gene flow between individuals, resulting in the formation of reproductively isolated lineages. Many years of experimental work involving crosses between closely-related species have shed light on the genetic basis of speciation and have identified the role of sex chromosomes in maintaining species barriers. However, these seminal experiments tell us more about their current reproductive isolation than the progression to this state. To shed light on this progression, the divergence history of the sex chromosome needs to be reconstructed and contrasted with the history of autosomes. Despite the accumulation of genome data, model-testing approaches are rarely used. In fact, genome scan studies are essentially descriptive, which limits our understanding of the evolutionary processes underlying observed molecular patterns. Another caveat of present speciation research is that it overlooks the diversity of sex chromosomes, and how different types of sex chromosomes (e.g. the X chromosome in mammals and the Z chromosome in birds) can affect the evolution of reproductive isolation. Here, I propose to evaluate the contribution of various types of sex chromosomes to the establishment of species barriers by using a model-based approach. Firstly, we will obtain theoretical predictions of the relative introgression rates of the sex chromosome and autosomes under different genetic architectures. This is a prerequisite to evaluate the potential of empirical work to identify differences in the molecular footprint of sex-linked versus autosomal introgression. Secondly, we will test through model fitting how much gene flow is impeded by the sex chromosome compared to autosomes. Such a unique statistical framework will be applied for the first time to various non-model taxa characterized by different types of sex chromosomes. Thirdly, we will investigate if the rate of speciation is affected by the different types of sex chromosome in a specific group (flies and mosquitoes), recently found to have undergone sex chromosome turnovers. This proposal will substantially advance our understanding of the role of sex chromosomes in speciation by combining theory and model fitting to next-generation sequencing data.

Speciation is a gradual process that requires the cessation of gene flow at a genome scale, ultimately producing reproductively isolated lineages. Two of the most consistent empirical patterns in speciation, stemming from experimental studies in Drosophila, suggest that the sex chromosome is critical for reproductive isolation. However, these seminal experiments tell us more about their current reproductive isolation than the progression to this state. To shed light on this progression, the divergence history of the sex chromosome needs to be reconstructed and contrasted with the history of autosomes. Moreover, we lack multilocus theoretical predictions for the rates of introgression of different chromosomes in various scenarios, which is critical to evaluate the potential of empirical studies to find differences in their introgression patterns. Another caveat of present speciation research is that it overlooks the diversity of sex chromosomes (e.g., the X chromosome in mammals and the Z chromosome in birds), and how different types of sex chromosomes can affect the evolution of reproductive isolation. Thus the goal of this project is to quantify the barrier effect played by sex chromosomes in relation to autosomes by comparing their rate of interspecific introgression using a modeling and genomic data analysis approach. Our most significant theoretical result is that the specific features of sex chromosomes (hemizygosity and absence of recombination in the heterogametic sex) lead to reduced levels of introgression on the X (or Z) compared to autosomes, but effects remain overall small. Therefore these effects may be too subtle to be observed in real data. To test these predictions, we developed a flexible statistical method based on an Approximate Bayesian Computation framework. This method easily scales to genomic data, and can accommodate complex divergence histories. It is implemented in a program called DILS and is also available via a web interface. Work in progress is to apply this model-testing approach on genomic data from various sexual systems. The estimation of interspecific introgression rates of the two types of chromosomes will provide evidence of whether or not the sex chromosome is a more efficient barrier to gene flow than autosomes in nature, and will shed light on the underlying causes. As a whole this project thus provides for the first time a quantitative vision of the role of sex chromosomes in speciation.