Unbroken: RADseq remains a powerful tool for understanding the genetics of adaptation in natural populations

Recently, Lowry et al. addressed the ability of RADseq approaches to detect loci under selection in genome scans. While the authors raise important considerations, such as accounting for the extent of linkage disequilibrium in a study system, we strongly disagree with their overall view of the ability of RADseq to inform our understanding of the genetic basis of adaptation. The family of RADseq protocols has radically improved the field of population genomics, expanding by several orders of magnitude the number of markers available while substantially reducing the cost per marker. Researchers whose goal is to identify regions of the genome under selection must consider the LD of the experimental system; however, there is no magical LD cutoff below which researchers should refuse to use RADseq. Lowry et al. further made two major arguments: a theoretical argument that modeled the likelihood of detecting selective sweeps with RAD markers, and gross summaries based on an anecdotal collection of RAD studies. Unfortunately, their simulations were off by two orders of magnitude in the worst case, while their anecdotes merely showed that it is possible to get widely divergent densities of RAD tags for any particular experiment, either by design or due to experimental efficacy. We strongly argue that RADseq remains a powerful and efficient approach that provides sufficient marker density for studying selection in many natural populations. Given limited resources, we argue that researchers should consider a wide range of trade‐offs among genomic techniques, in light of their study question and the power of different techniques to answer it.     

Parallel and nonparallel genome‐wide divergence among replicate population pairs of freshwater and anadromous Atlantic salmon

Little is known about the genetic basis differentiating resident and anadromous forms found in many salmonid species. Using a medium‐density SNP array, we documented genomic diversity and divergence at 2336 genetically mapped loci among three pairs of North American anadromous and freshwater Atlantic salmon populations. Our results show that across the genome, freshwater populations have lower diversity and a smaller proportion of private polymorphism relative to anadromous populations. Moreover, differentiation was more pronounced among freshwater than among anadromous populations at multiple spatial scales, suggesting a large effect of genetic drift in these isolated freshwater populations. Using nonhierarchical and hierarchical genome scans, we identified hundreds of markers spread across the genome that are potentially under divergent selection between anadromous and freshwater populations, but few outlier loci were repeatedly found in all three freshwater–anadromous comparisons. Similarly, a sliding window analysis revealed numerous regions of high divergence that were nonparallel among the three comparisons. These last results show little evidence for the parallel evolution of alleles selected for in freshwater populations, but suggest nonparallel adaptive divergence at many loci of small effects distributed through the genome. Overall, this study emphasizes the important role of genetic drift in driving genome‐wide reduction in diversity and divergence in freshwater Atlantic salmon populations and suggests a complex multigenic basis of adaptation to resident and anadromous strategies with little parallelism.     

Linkage disequilibrium network analysis (LDna) gives a global view of chromosomal inversions,local adaptation and geographic structure

Recent advances in sequencing allow population‐genomic data to be generated for virtually any species. However, approaches to analyse such data lag behind the ability to generate it, particularly in nonmodel species. Linkage disequilibrium (LD, the nonrandom association of alleles from different loci) is a highly sensitive indicator of many evolutionary phenomena including chromosomal inversions, local adaptation and geographical structure. Here, we present linkage disequilibrium network analysis (LDna), which accesses information on LD shared between multiple loci genomewide. In LD networks, vertices represent loci, and connections between vertices represent the LD between them. We analysed such networks in two test cases: a new restriction‐site‐associated DNA sequence (RAD‐seq) data set for Anopheles baimaii, a Southeast Asian malaria vector; and a well‐characterized single nucleotide polymorphism (SNP) data set from 21 three‐spined stickleback individuals. In each case, we readily identified five distinct LD network clusters (single‐outlier clusters, SOCs), each comprising many loci connected by high LD. In A. baimaii, further population‐genetic analyses supported the inference that each SOC corresponds to a large inversion, consistent with previous cytological studies. For sticklebacks, we inferred that each SOC was associated with a distinct evolutionary phenomenon: two chromosomal inversions, local adaptation, population‐demographic history and geographic structure. LDna is thus a useful exploratory tool, able to give a global overview of LD associated with diverse evolutionary phenomena and identify loci potentially involved. LDna does not require a linkage map or reference genome, so it is applicable to any population‐genomic data set, making it especially valuable for nonmodel species.     

Contrasting patterns of genome‐wide polymorphism in the native and invasive range of the marine mollusc Crepidula fornicata

Selection processes are believed to be an important evolutionary driver behind the successful establishment of nonindigenous species, for instance through adaptation for invasiveness (e.g. dispersal mechanisms and reproductive allocation). However, evidence supporting this assumption is still scarce. Genome scans have often identified loci with atypical patterns of genetic differentiation (i.e. outliers) indicative of selection processes. Using microsatellite‐ and AFLP‐based genome scans, we looked for evidence of selection following the introduction of the mollusc Crepidula fornicata. Native to the northwestern Atlantic, this gastropod has become an emblematic invader since its introduction during the 19th and 20th centuries in the northeastern Atlantic and northeastern Pacific. We examined 683 individuals from seven native and 15 introduced populations spanning the latitudinal introduction and native ranges of the species. Our results confirmed the previously documented high genetic diversity in native and introduced populations with little genetic structure between the two ranges, a pattern typical of marine invaders. Analysing 344 loci, no outliers were detected between the introduced and native populations or in the introduced range. The genomic sampling may have been insufficient to reveal selection especially if it acts on traits determined by a few genes. Eight outliers were, however, identified within the native range, underlining a genetic singularity congruent with a well‐known biogeographical break along the Florida. Our results call into question the relevance of AFLP genome scans in detecting adaptation on the timescale of biological invasions: genome scans often reveal long‐term adaptation involving numerous genes throughout the genome but seem less effective in detecting recent adaptation from pre‐existing variation on polygenic traits. This study advocates other methods to detect selection effects during biological invasions—for example on phenotypic traits, although genome scans may remain useful for elucidating introduction histories.     

RADseq approaches and applications for forest tree genetics

As tree species vary extensively in genome size, complexity, and resource development, reduced representation methods have been increasingly employed for the generation of population genomic data. By allowing rapid marker discovery and genotyping for thousands of genomic regions in many individuals without requiring genomic resources, restriction site-associated DNA sequencing (RADseq) methods have dramatically improved our ability to bring population genomic perspectives to non-model trees. The rapid recent increase in studies of trees utilizing RADseq suggests that it is likely to become among the most common approaches for generating genome-wide data for a variety of applications. Here we provide a practical review of RADseq and its application to research areas of tree genetics. We briefly review RADseq laboratory methods and consider analytical approaches for assembly, variant calling, and bioinformatic processing. To guide considerations for study design, we use in silico analyses of eight available tree genomes to illustrate how expected marker number and density vary across laboratory approaches and genome sizes, and to consider the ability of RADseq designs to query coding regions. We review the empirical use of RADseq for different research objectives, considering its strengths and limitations. Many studies have used RADseq data to perform genome scans for selection, although limited marker density and linkage disequilibrium will often compromise its utility for such analyses. Regardless of this limitation, RADseq offers a powerful and inexpensive technique for generating genome-wide SNP data that can greatly contribute to research spanning phylogenetic and population genetic inference, linkage mapping, and quantitative genetic parameter estimation for tree genetics.     

Tracking changes in chromosomal arrangements and their genetic content during adaptation

There is considerable evidence for an adaptive role of inversions, but how their genetic content evolves and affects the subsequent evolution of chromosomal polymorphism remains controversial. Here, we track how life‐history traits, chromosomal arrangements and 22 microsatellites, within and outside inversions, change in three replicated populations of Drosophila subobscura for 30 generations of laboratory evolution since founding from the wild. The dynamics of fitness‐related traits indicated adaptation to the new environment concomitant with directional evolution of chromosomal polymorphism. Evidence of selective changes in frequency of inversions was obtained for seven of 23 chromosomal arrangements, corroborating a role for inversions in adaptation. The evolution of linkage disequilibrium between some microsatellites and chromosomes suggested that adaptive changes in arrangements involved changes in their genetic content. Several microsatellite alleles increased in frequency more than expected by drift in targeted inversions in all replicate populations. In particular, there were signs of selection in the O₃₊₄ arrangement favouring a combination of alleles in two loci linked to the inversion and changing along with it, although the lack of linkage disequilibrium between these loci precludes epistatic selection. Seven other alleles increased in frequency within inversions more than expected by drift, but were not in linkage disequilibrium with them. Possibly these alleles were hitchhiking along with alleles under selection that were not specific to those inversions. Overall, the selection detected on the genetic content of inversions, despite limited coverage of the genome, suggests that genetic changes within inversions play an important role in adaptation.     

Restriction site‐associated DNA sequencing,genotyping error estimation and de novo assembly optimization for population genetic inference

Restriction site‐associated DNA sequencing (RADseq) provides researchers with the ability to record genetic polymorphism across thousands of loci for nonmodel organisms, potentially revolutionizing the field of molecular ecology. However, as with other genotyping methods, RADseq is prone to a number of sources of error that may have consequential effects for population genetic inferences, and these have received only limited attention in terms of the estimation and reporting of genotyping error rates. Here we use individual sample replicates, under the expectation of identical genotypes, to quantify genotyping error in the absence of a reference genome. We then use sample replicates to (i) optimize de novo assembly parameters within the program Stacks, by minimizing error and maximizing the retrieval of informative loci; and (ii) quantify error rates for loci, alleles and single‐nucleotide polymorphisms. As an empirical example, we use a double‐digest RAD data set of a nonmodel plant species, Berberis alpina, collected from high‐altitude mountains in Mexico.     

Responsible RAD: Striving for best practices in population genomic studies of adaptation

Two recent articles were written in response to our paper “Breaking RAD: An evaluation of the utility of restriction site associated DNA sequencing scans of adaptation.” While we agree with some of the comments made by the authors of these two response papers, we still believe caution should be employed in RADseq studies that aim to detect loci that contribute to adaptation. In this rebuttal, we evaluate the key points made in these papers, attempt to identify a middle ground and make suggestions for responsibly conducting future studies to understand the genomewide mechanisms of adaptation.     

Intrapopulation genomics in a model mutualist: Population structure and candidate symbiosis genes under selection in Medicago truncatula

Bottom‐up evolutionary approaches, including geographically explicit population genomic analyses, have the power to reveal the mechanistic basis of adaptation. Here, we conduct a population genomic analysis in the model legume, Medicago truncatula, to characterize population genetic structure and identify symbiosis‐related genes showing evidence of spatially variable selection. Using RAD‐seq, we generated over 26,000 SNPs from 191 accessions from within three regions of the native range in Europe. Results from STRUCTURE analysis identify five distinct genetic clusters with divisions that separate east and west regions in the Mediterranean basin. Much of the genetic variation is maintained within sampling sites, and there is evidence for isolation by distance. Extensive linkage disequilibrium was identified, particularly within populations. We conducted genetic outlier analysis with F_ST‐based genome scans and a Bayesian modeling approach (PCAdapt). There were 70 core outlier loci shared between these distinct methods with one clear candidate symbiosis related gene, DMI1. This work sets that stage for functional experiments to determine the important phenotypes that selection has acted upon and complementary efforts in rhizobium populations.     

Population genomic evidence for adaptive differentiation in the Baltic Sea herring

Detecting and estimating the degree of genetic differentiation among populations of highly mobile marine fish having pelagic larval stages is challenging because their effective population sizes can be large, and thus, little genetic drift and differentiation is expected in neutral genomic sites. However, genomic sites subject to directional selection stemming from variation in local environmental conditions can still show substantial genetic differentiation, yet these signatures can be hard to detect with low‐throughput approaches. Using a pooled RAD‐seq approach, we investigated genomewide patterns of genetic variability and differentiation within and among 20 populations of Atlantic herring in the Baltic Sea (and adjacent Atlantic sites), where previous low‐throughput studies and/or studies based on few populations have found limited evidence for genetic differentiation. Stringent quality control was applied in the filtering of 1 791 254 SNPs, resulting in a final data set of 68 182 polymorphic loci. Clear differentiation was identified between Atlantic and Baltic populations in many genomic sites, while differentiation within the Baltic Sea area was weaker and geographically less structured. However, outlier analyses – whether including all populations or only those within the Baltic Sea – uncovered hundreds of directionally selected loci in which variability was associated with either salinity, temperature or both. Hence, our results support the view that although the degree of genetic differentiation among Baltic Sea herring populations is low, there are many genomic regions showing elevated divergence, apparently as a response to temperature‐ and salinity‐related natural selection. As such, the results add to the increasing evidence of local adaptation in highly mobile marine organisms, and those in the young Baltic Sea in particular.     

Restricted gene flow and local adaptation highlight the vulnerability of high‐latitude reefs to rapid environmental change

Global climate change poses a serious threat to the future health of coral reef ecosystems. This calls for management strategies that are focused on maximizing the evolutionary potential of coral reefs. Fundamental to this is an accurate understanding of the spatial genetic structure in dominant reef‐building coral species. In this study, we apply a genotyping‐by‐sequencing approach to investigate genome‐wide patterns of genetic diversity, gene flow, and local adaptation in a reef‐building coral, Pocillopora damicornis, across 10 degrees of latitude and a transition from temperate to tropical waters. We identified strong patterns of differentiation and reduced genetic diversity in high‐latitude populations. In addition, genome‐wide scans for selection identified a number of outlier loci putatively under directional selection with homology to proteins previously known to be involved in heat tolerance in corals and associated with processes such as photoprotection, protein degradation, and immunity. This study provides genomic evidence for both restricted gene flow and local adaptation in a widely distributed coral species, and highlights the potential vulnerability of leading‐edge populations to rapid environmental change as they are locally adapted, reproductively isolated, and have reduced levels of genetic diversity.     

A microsatellite‐based linkage map for song sparrows (Melospiza melodia)

Although linkage maps are important tools in evolutionary biology, their availability for wild populations is limited. The population of song sparrows (Melospiza melodia) on Mandarte Island, Canada, is among the more intensively studied wild animal populations. Its long‐term pedigree data, together with extensive genetic sampling, have allowed the study of a range of questions in evolutionary biology and ecology. However, the availability of genetic markers has been limited. We here describe 191 new microsatellite loci, including 160 high‐quality polymorphic autosomal, 7 Z‐linked and 1 W‐linked markers. We used these markers to construct a linkage map for song sparrows with a total sex‐averaged map length of 1731 cM and covering 35 linkage groups, and hence, these markers cover most of the 38–40 chromosomes. Female and male map lengths did not differ significantly. We then bioinformatically mapped these loci to the zebra finch (Taeniopygia guttata) genome and found that linkage groups were conserved between song sparrows and zebra finches. Compared to the zebra finch, marker order within small linkage groups was well conserved, whereas the larger linkage groups showed some intrachromosomal rearrangements. Finally, we show that as expected, recombination frequency between linked loci explained the majority of variation in gametic phase disequilibrium. Yet, there was substantial overlap in gametic phase disequilibrium between pairs of linked and unlinked loci. Given that the microsatellites described here lie on 35 of the 38–40 chromosomes, these markers will be useful for studies in this species, as well as for comparative genomics studies with other species.     

The genomic signature of parallel adaptation from shared genetic variation

Parallel adaptation is common and may often occur from shared genetic variation, but the genomic consequences of this process remain poorly understood. We first use individual‐based simulations to demonstrate that comparisons between populations adapted in parallel to similar environments from shared variation reveal a characteristic genomic signature around a selected locus: a low‐divergence valley centred at the locus and flanked by twin peaks of high divergence. This signature is initiated by the hitchhiking of haplotype tracts differing between derived populations in the broader neighbourhood of the selected locus (driving the high‐divergence twin peaks) and shared haplotype tracts in the tight neighbourhood of the locus (driving the low‐divergence valley). This initial hitchhiking signature is reinforced over time because the selected locus acts as a barrier to gene flow from the source to the derived populations, thus promoting divergence by drift in its close neighbourhood. We next empirically confirm the peak‐valley‐peak signature by combining targeted and RAD sequence data at three candidate adaptation genes in multiple marine (source) and freshwater (derived) populations of threespine stickleback. Finally, we use a genome‐wide screen for the peak‐valley‐peak signature to discover additional genome regions involved in parallel marine‐freshwater divergence. Our findings offer a new explanation for heterogeneous genomic divergence and thus challenge the standard view that peaks in population divergence harbour divergently selected loci and that low‐divergence regions result from balancing selection or localized introgression. We anticipate that genome scans for peak‐valley‐peak divergence signatures will promote the discovery of adaptation genes in other organisms.     

Evidence for adaptation from standing genetic variation on an antimicrobial peptide gene in the mussel Mytilus edulis

Genome scans of population differentiation identify candidate loci for adaptation but provide little information on how selection has influenced the genetic structure of these loci. Following a genome scan, we investigated the nature of the selection responsible for the outlying differentiation observed between populations of the marine mussel Mytilus edulis at a leucine/arginine polymorphism (L31R) in the antimicrobial peptide MGD2. We analysed DNA sequence polymorphisms, allele frequencies and population differentiation of polymorphisms closely linked to L31R, and pairwise and third‐order linkage disequilibria. An outlying level of population differentiation was observed at L31R only, while no departure from panmixia was observed at linked loci surrounding L31R, as in most of the genome. Selection therefore seems to affect L31R directly. Three hypotheses can explain the lack of differentiation in the chromosomal region close to L31R: (i) hitchhiking has occurred but migration and recombination subsequently erased the signal, (ii) selection was weak enough and recombination strong enough to limit the hitchhiking effect to a very small chromosomal region or (iii) selection acted on a pre‐existing polymorphism (i.e. standing variation) at linkage equilibrium with its background. Linkage equilibrium was observed between L31R and linked polymorphisms in every population analysed, as expected under the three hypotheses. However, linkage disequilibrium was observed in some populations between pairs of loci located upstream and downstream to L31R, generating a complex pattern of third‐order linkage disequilibria which is best explained by the hypothesis of selection on a pre‐existing polymorphism. We hypothesise that selection could be either balanced, maintaining alleles at different frequencies depending on the pathogen community encountered locally by mussels, or intermittent, resulting in sporadic fluctuations in allele frequency.     

Combining molecular evolution and environmental genomics to unravel adaptive processes of MHC class IIB diversity in European minnows (Phoxinus phoxinus)

Host–pathogen interactions are a major evolutionary force promoting local adaptation. Genes of the major histocompatibility complex (MHC) represent unique candidates to investigate evolutionary processes driving local adaptation to parasite communities. The present study aimed at identifying the relative roles of neutral and adaptive processes driving the evolution of MHC class IIB (MHCIIB) genes in natural populations of European minnows (Phoxinus phoxinus). To this end, we isolated and genotyped exon 2 of two MHCIIB gene duplicates (DAB1 and DAB3) and 1′665 amplified fragment length polymorphism (AFLP) markers in nine populations, and characterized local bacterial communities by 16S rDNA barcoding using 454 amplicon sequencing. Both MHCIIB loci exhibited signs of historical balancing selection. Whereas genetic differentiation exceeded that of neutral markers at both loci, the populations' genetic diversities were positively correlated with local pathogen diversities only at DAB3. Overall, our results suggest pathogen‐mediated local adaptation in European minnows at both MHCIIB loci. While at DAB1 selection appears to favor different alleles among populations, this is only partially the case in DAB3, which appears to be locally adapted to pathogen communities in terms of genetic diversity. These results provide new insights into the importance of host–pathogen interactions in driving local adaptation in the European minnow, and highlight that the importance of adaptive processes driving MHCIIB gene evolution may differ among duplicates within species, presumably as a consequence of alternative selective regimes or different genomic context.     

SNPs across time and space: population genomic signatures of founder events and epizootics in the House Finch (Haemorhous mexicanus)

Identifying genomic signatures of natural selection can be challenging against a background of demographic changes such as bottlenecks and population expansions. Here, we disentangle the effects of demography from selection in the House Finch (Haemorhous mexicanus) using samples collected before and after a pathogen‐induced selection event. Using ddRADseq, we genotyped over 18,000 SNPs across the genome in native pre‐epizootic western US birds, introduced birds from Hawaii and the eastern United States, post‐epizootic eastern birds, and western birds sampled across a similar time span. We found 14% and 7% reductions in nucleotide diversity, respectively, in Hawaiian and pre‐epizootic eastern birds relative to pre‐epizootic western birds, as well as elevated levels of linkage disequilibrium and other signatures of founder events. Despite finding numerous significant frequency shifts (outlier loci) between pre‐epizootic native and introduced populations, we found no signal of reduced genetic diversity, elevated linkage disequilibrium, or outlier loci as a result of the epizootic. Simulations demonstrate that the proportion of outliers associated with founder events could be explained by genetic drift. This rare view of genetic evolution across time in an invasive species provides direct evidence that demographic shifts like founder events have genetic consequences more widespread across the genome than natural selection.     

Replicated analysis of the genetic architecture of quantitative traits in two wild great tit populations

Currently, there is much debate on the genetic architecture of quantitative traits in wild populations. Is trait variation influenced by many genes of small effect or by a few genes of major effect? Where is additive genetic variation located in the genome? Do the same loci cause similar phenotypic variation in different populations? Great tits (Parus major) have been studied extensively in long‐term studies across Europe and consequently are considered an ecological ‘model organism’. Recently, genomic resources have been developed for the great tit, including a custom SNP chip and genetic linkage map. In this study, we used a suite of approaches to investigate the genetic architecture of eight quantitative traits in two long‐term study populations of great tits—one in the Netherlands and the other in the United Kingdom. Overall, we found little evidence for the presence of genes of large effects in either population. Instead, traits appeared to be influenced by many genes of small effect, with conservative estimates of the number of contributing loci ranging from 31 to 310. Despite concordance between population‐specific heritabilities, we found no evidence for the presence of loci having similar effects in both populations. While population‐specific genetic architectures are possible, an undetected shared architecture cannot be rejected because of limited power to map loci of small and moderate effects. This study is one of few examples of genetic architecture analysis in replicated wild populations and highlights some of the challenges and limitations researchers will face when attempting similar molecular quantitative genetic studies in free‐living populations.     

Population genomics of natural and experimental populations of guppies (Poecilia reticulata)

Convergent evolution represents one of the best lines of evidence for adaptation, but few cases of phenotypic convergence are understood at the genetic level. Guppies inhabiting the Northern Mountain Range of Trinidad provide a classic example of phenotypic convergent evolution, where adaptation to low or high predation environments has been found for a variety of traits. A major advantage of this system is the possibility of long‐term experimental studies in nature, including transplantation from high to low predation sites. We used genome scans of guppies from three natural high and low predation populations and from two experimentally established populations and their sources to examine whether phenotypic convergent evolution leaves footprints at the genome level. We used population‐genetic modelling approaches to reconstruct the demographic history and migration among sampled populations. Naturally colonized low predation populations had signatures of increased effective population size since colonization, while introduction populations had signatures of decreased effective population size. Only a small number of regions across the genome had signatures of selection in all natural populations. However, the two experimental populations shared many genomic regions under apparent selection, more than expected by chance. This overlap coupled with a population decrease since introduction provides evidence for convergent selection occurring in the two introduced populations. The lack of genetic convergence in the natural populations suggests that convergent evolution is lacking in these populations or that the effects of selection become difficult to detect after a long‐time period.     

AFLP genome scan to detect genetic structure and candidate loci under selection for local adaptation of the invasive weed Mikania micrantha

Why some species become successful invaders is an important issue in invasive biology. However, limited genomic resources make it very difficult for identifying candidate genes involved in invasiveness. Mikania micrantha H.B.K. (Asteraceae), one of the world's most invasive weeds, has adapted rapidly in response to novel environments since its introduction to southern China. In its genome, we expect to find outlier loci under selection for local adaptation, critical to dissecting the molecular mechanisms of invasiveness. An explorative amplified fragment length polymorphism (AFLP) genome scan was used to detect candidate loci under selection in 28 M. micrantha populations across its entire introduced range in southern China. We also estimated population genetic parameters, bottleneck signatures, and linkage disequilibrium. In binary characters, such as presence or absence of AFLP bands, if all four character combinations are present, it is referred to as a character incompatibility. Since character incompatibility is deemed to be rare in populations with extensive asexual reproduction, a character incompatibility analysis was also performed in order to infer the predominant mating system in the introduced M. micrantha populations. Out of 483 AFLP loci examined using stringent significance criteria, 14 highly credible outlier loci were identified by Dfdist and Bayescan. Moreover, remarkable genetic variation, multiple introductions, substantial bottlenecks and character compatibility were found to occur in M. micrantha. Thus local adaptation at the genome level indeed exists in M. micrantha, and may represent a major evolutionary mechanism of successful invasion. Interactions between genetic diversity, multiple introductions, and reproductive modes contribute to increase the capacity of adaptive evolution.