How to identify sex chromosomes and their turnover

Abstract

Although sex is a fundamental component of eukaryotic reproduction, the genetic systems that control sex determination are highly variable. In many organisms the presence of sex chromosomes is associated with female or male development. Although certain groups possess stable and conserved sex chromosomes, others exhibit rapid sex chromosome evolution, including transitions between male and female heterogamety, and turnover in the chromosome pair recruited to determine sex. These turnover events have important consequences for multiple facets of evolution, as sex chromosomes are predicted to play a central role in adaptation, sexual dimorphism, and speciation. However, our understanding of the processes driving the formation and turnover of sex chromosome systems is limited, in part because we lack a complete understanding of interspecific variation in the mechanisms by which sex is determined. New bioinformatic methods are making it possible to identify and characterize sex chromosomes in a diverse array of non-model species, rapidly filling in the numerous gaps in our knowledge of sex chromosome systems across the tree of life. In turn, this growing data set is facilitating and fueling efforts to address many of the unanswered questions in sex chromosome evolution. Here, we synthesize the available bioinformatic approaches to produce a guide for characterizing sex chromosome system and identity simultaneously across clades of organisms. Furthermore, we survey our current understanding of the processes driving sex chromosome turnover, and highlight important avenues for future research.

Keywords: bioinformatics; next-generation sequencing; sex chromosome turnover; sex chromosomes.

Figure 1

Illustration of the homomorphic‐heteromorphic sex chromosome continuum. Sex chromosomes can range from heteromorphic, where the X and Y (or Z and W) chromosomes are diverged and highly distinct, to homomorphic, where pairs are nearly identical in gene content and size. However, sex chromosomes can vary in their degree of sequence differentiation not just among species (top panel) but also among strata within a species (bottom panel). Strata are regions of the chromosome where recombination between the sex chromosomes has been halted independently and therefore are of different ages. Different methods for identifying sex‐linked loci will be appropriate for species/strata at different points on this continuum. Purple scale indicates sequence differentiation between chromosomes or strata, where lighter purple shows greater divergence

Figure 2

Overview of bioinformatic methods available for sex chromosome identification. This figure is based on XY sex chromosomes, but all methods can be inverted for ZW systems. Top left panel shows the key. Top right panel solid bars show which methods are most effective along different points of the sex chromosome divergence continuum. Dashed bar indicates that the method is partially effective. (a) Genomic coverage approach: in nonrecombining regions of sex chromosomes, where the Y has degenerated, males have only one X chromosome, and thus show a reduced genomic coverage relative to females. (b) Expression‐based approach: male RNA‐seq reads are mapped to a female reference. Unmapped reads are assembled into de novo contigs to identify putative Y‐linked sequences. Re‐mapping female transcripts to these contigs can be used to verify male‐limitation. (c) Association‐based approach: male and female RAD‐tags are compared to isolate male‐specific RAD loci. (d) SNP density approach: in younger regions of the sex chromosomes, which still retain high sequence similarity between the X and the Y, we expect an increase in male SNP density compared to females, as Y reads, carrying Y‐specific SNPs, still map to the homologous X regions. This SNP density pattern is not expected in old strata with substantial Y degeneration, as the X is effectively hemizygous in males. Contrasting sex differences in coverage and SNP density is a powerful approach to identify sex‐linked regions. (e) Segregation analysis approach: SNP data obtained from parents and progeny are analyzed in a statistical framework to assess the likelihood of autosomal versus sex‐linked segregation patterns. (f) Linkage mapping approach: recombination patterns of parents and offspring are compared, and regions with no recombination between males and females indicate putative sex‐linked regions