A long evolutionary reach for fishing nets.

Earlier maturation of Atlantic salmon is linked to indirect effects of fisheries on its prey.

Batch effects in population genomic studies with low-coverage whole genome sequencing data: Causes, detection and mitigation.

Over the past few decades, there has been an explosion in the amount of publicly available sequencing data. This opens new opportunities for combining data sets to achieve unprecedented sample sizes, spatial coverage or temporal replication in population genomic studies. However, a common concern is that nonbiological differences between data sets may generate patterns of variation in the data that can confound real biological patterns, a problem known as batch effects. In this paper, we compare two batches of low-coverage whole genome sequencing (lcWGS) data generated from the same populations of Atlantic cod (Gadus morhua). First, we show that with a "batch-effect-naive" bioinformatic pipeline, batch effects systematically biased our genetic diversity estimates, population structure inference and selection scans. We then demonstrate that these batch effects resulted from multiple technical differences between our data sets, including the sequencing chemistry (four-channel vs. two-channel), sequencing run, read type (single-end vs. paired-end), read length (125 vs. 150 bp), DNA degradation level (degraded vs. well preserved) and sequencing depth (0.8× vs. 0.3× on average). Lastly, we illustrate that a set of simple bioinformatic strategies (such as different read trimming and single nucleotide polymorphism filtering) can be used to detect batch effects in our data and substantially mitigate their impact. We conclude that combining data sets remains a powerful approach as long as batch effects are explicitly accounted for. We focus on lcWGS data in this paper, which may be particularly vulnerable to certain causes of batch effects, but many of our conclusions also apply to other sequencing strategies.

A beginner's guide to low-coverage whole genome sequencing for population genomics.

Low-coverage whole genome sequencing (lcWGS) has emerged as a powerful and cost-effective approach for population genomic studies in both model and nonmodel species. However, with read depths too low to confidently call individual genotypes, lcWGS requires specialized analysis tools that explicitly account for genotype uncertainty. A growing number of such tools have become available, but it can be difficult to get an overview of what types of analyses can be performed reliably with lcWGS data, and how the distribution of sequencing effort between the number of samples analysed and per-sample sequencing depths affects inference accuracy. In this introductory guide to lcWGS, we first illustrate how the per-sample cost for lcWGS is now comparable to RAD-seq and Pool-seq in many systems. We then provide an overview of software packages that explicitly account for genotype uncertainty in different types of population genomic inference. Next, we use both simulated and empirical data to assess the accuracy of allele frequency, genetic diversity, and linkage disequilibrium estimation, detection of population structure, and selection scans under different sequencing strategies. Our results show that spreading a given amount of sequencing effort across more samples with lower depth per sample consistently improves the accuracy of most types of inference, with a few notable exceptions. Finally, we assess the potential for using imputation to bolster inference from lcWGS data in nonmodel species, and discuss current limitations and future perspectives for lcWGS-based population genomics research. With this overview, we hope to make lcWGS more approachable and stimulate its broader adoption.

Chromosome-Level Assembly of the Atlantic Silverside Genome Reveals Extreme Levels of Sequence Diversity and Structural Genetic Variation.

The levels and distribution of standing genetic variation in a genome can provide a wealth of insights about the adaptive potential, demographic history, and genome structure of a population or species. As structural variants are increasingly associated with traits important for adaptation and speciation, investigating both sequence and structural variation is essential for wholly tapping this potential. Using a combination of shotgun sequencing, 10x Genomics linked reads and proximity-ligation data (Chicago and Hi-C), we produced and annotated a chromosome-level genome assembly for the Atlantic silverside (Menidia menidia)-an established ecological model for studying the phenotypic effects of natural and artificial selection-and examined patterns of genomic variation across two individuals sampled from different populations with divergent local adaptations. Levels of diversity varied substantially across each chromosome, consistently being highly elevated near the ends (presumably near telomeric regions) and dipping to near zero around putative centromeres. Overall, our estimate of the genome-wide average heterozygosity in the Atlantic silverside is among the highest reported for a fish, or any vertebrate (1.32-1.76% depending on inference method and sample). Furthermore, we also found extreme levels of structural variation, affecting ∼23% of the total genome sequence, including multiple large inversions (> 1 Mb and up to 12.6 Mb) associated with previously identified haploblocks showing strong differentiation between locally adapted populations. These extreme levels of standing genetic variation are likely associated with large effective population sizes and may help explain the remarkable adaptive divergence among populations of the Atlantic silverside.

Footprints of local adaptation span hundreds of linked genes in the Atlantic silverside genome.

The study of local adaptation in the presence of ongoing gene flow is the study of natural selection in action, revealing the functional genetic diversity most relevant to contemporary pressures. In addition to individual genes, genome-wide architecture can itself evolve to enable adaptation. Distributed across a steep thermal gradient along the east coast of North America, Atlantic silversides (Menidia menidia) exhibit an extraordinary degree of local adaptation in a suite of traits, and the capacity for rapid adaptation from standing genetic variation, but we know little about the patterns of genomic variation across the species range that enable this remarkable adaptability. Here, we use low-coverage, whole-transcriptome sequencing of Atlantic silversides sampled along an environmental cline to show marked signatures of divergent selection across a gradient of neutral differentiation. Atlantic silversides sampled across 1371 km of the southern section of its distribution have very low genome-wide differentiation (median F ST = 0.006 across 1.9 million variants), consistent with historical connectivity and observations of recent migrants. Yet almost 14,000 single nucleotide polymorphisms (SNPs) are nearly fixed (F ST > 0.95) for alternate alleles. Highly differentiated SNPs cluster into four tight linkage disequilibrium (LD) blocks that span hundreds of genes and several megabases. Variants in these LD blocks are disproportionately nonsynonymous and concentrated in genes enriched for multiple functions related to known adaptations in silversides, including variation in lipid storage, metabolic rate, and spawning behavior. Elevated levels of absolute divergence and demographic modeling suggest selection maintaining divergence across these blocks under gene flow. These findings represent an extreme case of heterogeneity in levels of differentiation across the genome, and highlight how gene flow shapes genomic architecture in continuous populations. Locally adapted alleles may be common features of populations distributed along environmental gradients, and will likely be key to conserving variation to enable future responses to environmental change.

Building a global genomics observatory: Using GEOME (the Genomic Observatories Metadatabase) to expedite and improve deposition and retrieval of genetic data and metadata for biodiversity research.

Genetic data represent a relatively new frontier for our understanding of global biodiversity. Ideally, such data should include both organismal DNA-based genotypes and the ecological context where the organisms were sampled. Yet most tools and standards for data deposition focus exclusively either on genetic or ecological attributes. The Genomic Observatories Metadatabase (GEOME: geome-db.org) provides an intuitive solution for maintaining links between genetic data sets stored by the International Nucleotide Sequence Database Collaboration (INSDC) and their associated ecological metadata. GEOME facilitates the deposition of raw genetic data to INSDCs sequence read archive (SRA) while maintaining persistent links to standards-compliant ecological metadata held in the GEOME database. This approach facilitates findable, accessible, interoperable and reusable data archival practices. Moreover, GEOME enables data management solutions for large collaborative groups and expedites batch retrieval of genetic data from the SRA. The article that follows describes how GEOME can enable genuinely open data workflows for researchers in the field of molecular ecology.

A comprehensive non-redundant reference transcriptome for the Atlantic silverside Menidia menidia.

The Atlantic silverside (Menidia menidia) has been the focus of extensive research efforts in ecology, evolutionary biology, and physiology over the past three decades, but lack of genomic resources has so far hindered examination of the molecular basis underlying the remarkable patterns of phenotypic variation described in this species. We here present the first reference transcriptome for M. menidia. We sought to capture a single representative sequence from as many genes as possible by first using a combination of Trinity and the CLC Genomics Workbench to de novo assemble contigs based on RNA-seq data from multiple individuals, tissue types, and life stages. To reduce redundancy, we passed the combined raw assemblies through a stringent filtering pipeline based both on sequence similarity to related species and computational predictions of transcript quality, condensing an initial set of >480,000 contigs to a final set of 20,998 representative contigs, amounting to a total length of 53.3 Mb. In this final assembly, 91% of the contigs were functionally annotated with putative gene function and gene ontology (GO) terms and/or InterProScan identifiers. The assembly contains complete or nearly complete copies of >95% of 248 highly conserved core genes present in low copy number across higher eukaryotes, and partial copies of another 3.8%, suggesting that our assembly provides relatively comprehensive coverage of the M. menidia transcriptome. The assembly provided here will be an important resource for future research.

Inference of natural selection from ancient DNA.

Evolutionary processes, including selection, can be indirectly inferred based on patterns of genomic variation among contemporary populations or species. However, this often requires unrealistic assumptions of ancestral demography and selective regimes. Sequencing ancient DNA from temporally spaced samples can inform about past selection processes, as time series data allow direct quantification of population parameters collected before, during, and after genetic changes driven by selection. In this Comment and Opinion, we advocate for the inclusion of temporal sampling and the generation of paleogenomic datasets in evolutionary biology, and highlight some of the recent advances that have yet to be broadly applied by evolutionary biologists. In doing so, we consider the expected signatures of balancing, purifying, and positive selection in time series data, and detail how this can advance our understanding of the chronology and tempo of genomic change driven by selection. However, we also recognize the limitations of such data, which can suffer from postmortem damage, fragmentation, low coverage, and typically low sample size. We therefore highlight the many assumptions and considerations associated with analyzing paleogenomic data and the assumptions associated with analytical methods.

Excavating ghost footprints and tangled trees from modern genomes.

Due to pervasive gene flow and admixture, simple bifurcating trees often do not provide an accurate representation of relationships among diverging lineages, but limited resolution in the available genomic data and the spatial distribution of samples has hindered detailed insights regarding the evolutionary and demographic history of many species and populations. In this issue of Molecular Ecology, Foote et al. (2019) combine a powerful sampling design with novel analytical methods adopted from human genetics to describe previously unrecognized patterns of recurrent vicariance and admixture among lineages in the globally distributed killer whale (Orcinus orca). Based on sequence data from modern samples alone, they discover clear signatures of ancient admixture with a now extinct "ghost" lineage, providing one of the first accounts of archaic introgression in a nonhominid species. Coupling a cost-effective sequencing strategy with novel analytical approaches, their paper provides a roadmap for advancing inference of evolutionary history in other nonmodel species, promising exciting times ahead for our field.

Practical low-coverage genomewide sequencing of hundreds of individually barcoded samples for population and evolutionary genomics in nonmodel species.

Today most population genomic studies of nonmodel organisms either sequence a subset of the genome deeply in each individual or sequence pools of unlabelled individuals. With a step-by-step workflow, we illustrate how low-coverage whole-genome sequencing of hundreds of individually barcoded samples is now a practical alternative strategy for obtaining genomewide data on a population scale. We used a highly efficient protocol to generate high-quality libraries for ~6.5 USD from each of 876 Atlantic silversides (a teleost fish with a genome size ~730 Mb) that we sequenced to 1-4× genome coverage. In the absence of a reference genome, we developed a bioinformatic pipeline for mapping the genomic reads to a de novo assembled reference transcriptome. This provides an 'in silico' method for exome capture that avoids the complexities and expenses of using wet chemistry for target isolation. Using novel tools for analysis of low-coverage data, we extracted population allele frequencies, individual genotype likelihoods and polymorphism data for 2 504 335 SNPs across the exome for the 876 fish. To illustrate the use of the resulting data, we present a preliminary analysis of geographical patterns in the exome data and a comparison of complete mitochondrial genome sequences for each individual (constructed from the low-coverage data) that show population colonization patterns along the US east coast. With a total cost per sample of less than 50 USD (including sequencing) and ability to prepare 96 libraries in only 5 h, our approach adds a viable new option to the population genomics toolbox.

The life aquatic: advances in marine vertebrate genomics.

The ocean is hypothesized to be where life on earth originated, and subsequent evolutionary transitions between marine and terrestrial environments have been key events in the origin of contemporary biodiversity. Here, we review how comparative genomic approaches are an increasingly important aspect of understanding evolutionary processes, such as physiological and morphological adaptation to the diverse habitats within the marine environment. In addition, we highlight how population genomics has provided unprecedented resolution for population structuring, speciation and adaptation in marine environments, which can have a low cost of dispersal and few physical barriers to gene flow, and can thus support large populations. Building upon this work, we outline the applications of genomics tools to conservation and their relevance to assessing the wide-ranging impact of fisheries and climate change on marine species.

Archived DNA reveals fisheries and climate induced collapse of a major fishery.

Fishing and climate change impact the demography of marine fishes, but it is generally ignored that many species are made up of genetically distinct locally adapted populations that may show idiosyncratic responses to environmental and anthropogenic pressures. Here, we track 80 years of Atlantic cod (Gadus morhua) population dynamics in West Greenland using DNA from archived otoliths in combination with fish population and niche based modeling. We document how the interacting effects of climate change and high fishing pressure lead to dramatic spatiotemporal changes in the proportions and abundance of different genetic populations, and eventually drove the cod fishery to a collapse in the early 1970s. Our results highlight the relevance of fisheries management at the level of genetic populations under future scenarios of climate change.

Population genomics of marine fishes: next-generation prospects and challenges.

Over the past few years, technological advances have facilitated giant leaps forward in our ability to generate genome-wide molecular data, offering exciting opportunities for gaining new insights into the ecology and evolution of species where genomic information is still limited. Marine fishes are valuable organisms for advancing our understanding of evolution on historical and contemporary time scales, and here we highlight areas in which research on these species is likely to be particularly important in the near future. These include possibilities for gaining insights into processes on ecological time scales, identifying genomic signatures associated with population divergence under gene flow, and determining the genetic basis of phenotypic traits. We also consider future challenges pertaining to the implementation of genome-wide coverage through next-generation sequencing and genotyping methods in marine fishes. Complications associated with fast decay of linkage disequilibrium, as expected for species with large effective population sizes, and the possibility that adaptation is associated with both soft selective sweeps and polygenic selection, leaving complex genomic signatures in natural populations, are likely to challenge future studies. However, the combination of high genome coverage and new statistical developments offers promising solutions. Thus, the next generation of studies is likely to truly facilitate the transition from population genetics to population genomics in marine fishes. This transition will advance our understanding of basic evolutionary processes and will offer new possibilities for conservation and management of valuable marine resources.

The use of archived tags in retrospective genetic analysis of fish.

Collections of historical tissue samples from fish (e.g. scales and otoliths) stored in museums and fisheries institutions are precious sources of DNA for conducting retrospective genetic analysis. However, in some cases, only external tags used for documentation of spatial dynamics of fish populations have been preserved. Here, we test the usefulness of fish tags as a source of DNA for genetic analysis. We extract DNA from historical tags from cod collected in Greenlandic waters between 1950 and 1968. We show that the quantity and quality of DNA recovered from tags is comparable to DNA from archived otoliths from the same individuals. Surprisingly, levels of cross-contamination do not seem to be significantly higher in DNA from external (tag) than internal (otolith) sources. Our study therefore demonstrates that historical tags can be a highly valuable source of DNA for retrospective genetic analysis of fish.

Spatiotemporal SNP analysis reveals pronounced biocomplexity at the northern range margin of Atlantic cod Gadus morhua.

Accurate prediction of species distribution shifts in the face of climate change requires a sound understanding of population diversity and local adaptations. Previous modeling has suggested that global warming will lead to increased abundance of Atlantic cod (Gadus morhua) in the ocean around Greenland, but the dynamics of earlier abundance fluctuations are not well understood. We applied a retrospective spatiotemporal population genomics approach to examine the temporal stability of cod population structure in this region and to search for signatures of divergent selection over a 78-year period spanning major demographic changes. Analyzing >900 gene-associated single nucleotide polymorphisms in 847 individuals, we identified four genetically distinct groups that exhibited varying spatial distributions with considerable overlap and mixture. The genetic composition had remained stable over decades at some spawning grounds, whereas complete population replacement was evident at others. Observations of elevated differentiation in certain genomic regions are consistent with adaptive divergence between the groups, indicating that they may respond differently to environmental variation. Significantly increased temporal changes at a subset of loci also suggest that adaptation may be ongoing. These findings illustrate the power of spatiotemporal population genomics for revealing biocomplexity in both space and time and for informing future fisheries management and conservation efforts.

The simple fool's guide to population genomics via RNA-Seq: an introduction to high-throughput sequencing data analysis.

High-throughput sequencing technologies are currently revolutionizing the field of biology and medicine, yet bioinformatic challenges in analysing very large data sets have slowed the adoption of these technologies by the community of population biologists. We introduce the 'Simple Fool's Guide to Population Genomics via RNA-seq' (SFG), a document intended to serve as an easy-to-follow protocol, walking a user through one example of high-throughput sequencing data analysis of nonmodel organisms. It is by no means an exhaustive protocol, but rather serves as an introduction to the bioinformatic methods used in population genomics, enabling a user to gain familiarity with basic analysis steps. The SFG consists of two parts. This document summarizes the steps needed and lays out the basic themes for each and a simple approach to follow. The second document is the full SFG, publicly available at http://sfg.stanford.edu, that includes detailed protocols for data processing and analysis, along with a repository of custom-made scripts and sample files. Steps included in the SFG range from tissue collection to de novo assembly, blast annotation, alignment, gene expression, functional enrichment, SNP detection, principal components and F(ST) outlier analyses. Although the technical aspects of population genomics are changing very quickly, our hope is that this document will help population biologists with little to no background in high-throughput sequencing and bioinformatics to more quickly adopt these new techniques.