Summary Restriction site‐Associated DNA sequencing (RAD‐seq) has become a widely adopted method for genotyping populations of model and non‐model organisms. Generating a reliable set of loci for downstream analysis requires appropriate use of bioinformatics software, such as the program stacks. Using three empirical RAD‐seq datasets, we demonstrate a method for optimising a de novo assembly of loci using stacks. By iterating values of the program's main parameters and plotting resultant core metrics for visualisation, researchers can gain a much better understanding of their dataset and select an optimal set of parameters; we present the 80% rule as a generally effective method to select the core parameters for stacks. Visualisation of the metrics plotted for the three RAD‐seq datasets shows that they differ in the optimal parameters that should be used to maximise the amount of available biological information. We also demonstrate that building loci de novo and then integrating alignment positions is more effective than aligning raw reads directly to a reference genome. Our methods will help the community in honing the analytical skills necessary to accurately assemble a RAD‐seq dataset.

Theory predicts that the sexes can achieve greater fitness if loci with sexually antagonistic polymorphisms become linked to the sex determining loci, and this can favour the spread of reduced recombination around sex determining regions. Given that sex-linked regions are frequently repetitive and highly heterozygous, few complete Y chromosome assemblies are available to test these ideas. The guppy system (Poecilia reticulata) has long been invoked as an example of sex chromosome formation resulting from sexual conflict. Early genetics studies revealed that male colour patterning genes are mostly but not entirely Y-linked, and that X-linkage may be most common in low predation populations. More recent population genomic studies of guppies have reached varying conclusions about the size and placement of the Y-linked region. However, this previous work used a reference genome assembled from short-read sequences from a female guppy. Here, we present a new guppy reference genome assembly from a male, using long-read PacBio single-molecule real-time sequencing (SMRT) and chromosome contact information. Our new assembly sequences across repeat- and GC-rich regions and thus closes gaps and corrects mis-assemblies found in the short-read female-derived guppy genome. Using this improved reference genome, we then employed broad population sampling to detect sex differences across the genome. We identified two small regions that showed consistent male-specific signals. Moreover, our results help reconcile the contradictory conclusions put forth by past population genomic studies of the guppy sex chromosome. Our results are consistent with a small Y-specific region and rare recombination in male guppies.

Humans have exploited the earth's metal resources for thousands of years leaving behind a legacy of toxic metal contamination and poor water quality. The southwest of England provides a well-defined example, with a rich history of metal mining dating to the Bronze Age. Mine water washout continues to negatively impact water quality across the region where brown trout (Salmo trutta L.) populations exist in both metal-impacted and relatively clean rivers. We used microsatellites to assess the genetic impact of mining practices on trout populations in this region. Our analyses demonstrated that metal-impacted trout populations have low genetic diversity and have experienced severe population declines. Metal-river trout populations are genetically distinct from clean-river populations, and also from one another, despite being geographically proximate. Using approximate Bayesian computation (ABC), we dated the origins of these genetic patterns to periods of intensive mining activity. The historical split of contemporary metal-impacted populations from clean-river fish dated to the Medieval period. Moreover, we observed two distinct genetic populations of trout within a single catchment and dated their divergence to the Industrial Revolution. Our investigation thus provides an evaluation of contemporary population genetics in showing how human-altered landscapes can change the genetic makeup of a species.

Studies of convergence in wild populations have been instrumental in understanding adaptation by providing strong evidence for natural selection. At the genetic level, we are beginning to appreciate that the re-use of the same genes in adaptation occurs through different mechanisms and can be constrained by underlying trait architectures and demographic characteristics of natural populations. Here, we explore these processes in naturally adapted high- (HP) and low-predation (LP) populations of the Trinidadian guppy, Poecilia reticulata. As a model for phenotypic change this system provided some of the earliest evidence of rapid and repeatable evolution in vertebrates; the genetic basis of which has yet to be studied at the whole-genome level. We collected whole-genome sequencing data from ten populations (176 individuals) representing five independent HP-LP river pairs across the three main drainages in Northern Trinidad. We evaluate population structure, uncovering several LP bottlenecks and variable between-river introgression that can lead to constraints on the sharing of adaptive variation between populations. Consequently, we found limited selection on common genes or loci across all drainages. Using a pathway type analysis, however, we find evidence of repeated selection on different genes involved in cadherin signaling. Finally, we found a large repeatedly selected haplotype on chromosome 20 in three rivers from the same drainage. Taken together, despite limited sharing of adaptive variation among rivers, we found evidence of convergent evolution associated with HP-LP environments in pathways across divergent drainages and at a previously unreported candidate haplotype within a drainage.

Studies of convergence in wild populations have been instrumental in understanding adaptation by providing strong evidence for natural selection. At the genetic level, we are beginning to appreciate that the re-use of the same genes in adaptation occurs through different mechanisms and can be constrained by underlying trait architectures and demographic characteristics of natural populations. Here, we explore these processes in naturally adapted high- (HP) and low-predation (LP) populations of the Trinidadian guppy, Poecilia reticulata. As a model for phenotypic change this system provided some of the earliest evidence of rapid and repeatable evolution in vertebrates; the genetic basis of which has yet to be studied at the whole-genome level. We collected whole-genome sequencing data from ten populations (176 individuals) representing five independent HP-LP river pairs across the three main drainages in Northern Trinidad. We evaluate population structure, uncovering several LP bottlenecks and variable between-river introgression that can lead to constraints on the sharing of adaptive variation between populations. Consequently, we found limited selection on common genes or loci across all drainages. Using a pathway type analysis, however, we find evidence of repeated selection on different genes involved in cadherin signalling. Finally, we found a large repeatedly selected haplotype on chromosome 20 in three rivers from the same drainage. Taken together, despite limited sharing of adaptive variation among rivers, we found evidence of convergent evolution associated with HP-LP environments in pathways across divergent drainages and at a previously unreported candidate haplotype within a drainage.