Resolving postglacial phylogeography using high-throughput sequencing

Two different high throughput sequencing approaches identify thousands of de novo genomic markers for the genetically depleted Bornean elephant

High throughput sequencing technologies are being applied to an increasing number of model species with a high-quality reference genome. The application and analyses of whole-genome sequence data in non-model species with no prior genomic information are currently under way. Recent sequencing technologies provide new opportunities for gathering genomic data in natural populations, laying the empirical foundation for future research in the field of conservation and population genomics. Here we present the case study of the Bornean elephant, which is the most endangered subspecies of Asian elephant and exhibits very low genetic diversity. We used two different sequencing platforms, the Roche 454 FLX (shotgun) and Illumina, GAIIx (Restriction site associated DNA, RAD) to evaluate the feasibility of the two methodologies for the discovery of de novo markers (single nucleotide polymorphism, SNPs and microsatellites) using low coverage data. Approximately, 6,683 (shotgun) and 14,724 (RAD) SNPs were detected within our elephant sequence dataset. Genotyping of a representative sample of 194 SNPs resulted in a SNP validation rate of ∼ 83 to 94% and 17% of the loci were polymorphic with a low diversity (H_o = 0.057). Different numbers of microsatellites were identified through shotgun (27,226) and RAD (868) techniques. Out of all di-, tri-, and tetra-microsatellite loci, 1,706 loci had sufficient flanking regions (shotgun) while only 7 were found with RAD. All microsatellites were monomorphic in the Bornean but polymorphic in another elephant subspecies. Despite using different sample sizes, and the well known differences in the two platforms used regarding sequence length and throughput, the two approaches showed high validation rate. The approaches used here for marker development in a threatened species demonstrate the utility of high throughput sequencing technologies as a starting point for the development of genomic tools in a non-model species and in particular for a species with low genetic diversity.

Rank and order: evaluating the performance of SNPs for individual assignment in a non-model organism

Single nucleotide polymorphisms (SNPs) are valuable tools for ecological and evolutionary studies. In non-model species, the use of SNPs has been limited by the number of markers available. However, new technologies and decreasing technology costs have facilitated the discovery of a constantly increasing number of SNPs. With hundreds or thousands of SNPs potentially available, there is interest in comparing and developing methods for evaluating SNPs to create panels of high-throughput assays that are customized for performance, research questions, and resources. Here we use five different methods to rank 43 new SNPs and 71 previously published SNPs for sockeye salmon: F(ST), informativeness (I(n)), average contribution to principal components (LC), and the locus-ranking programs BELS and WHICHLOCI. We then tested the performance of these different ranking methods by creating 48- and 96-SNP panels of the top-ranked loci for each method and used empirical and simulated data to obtain the probability of assigning individuals to the correct population using each panel. All 96-SNP panels performed similarly and better than the 48-SNP panels except for the 96-SNP BELS panel. Among the 48-SNP panels, panels created from F(ST), I(n), and LC ranks performed better than panels formed using the top-ranked loci from the programs BELS and WHICHLOCI. The application of ranking methods to optimize panel performance will become more important as more high-throughput assays become available.

Special features of RAD Sequencing data: implications for genotyping

Restriction site-associated DNA Sequencing (RAD-Seq) is an economical and efficient method for SNP discovery and genotyping. As with other sequencing-by-synthesis methods, RAD-Seq produces stochastic count data and requires sensitive analysis to develop or genotype markers accurately. We show that there are several sources of bias specific to RAD-Seq that are not explicitly addressed by current genotyping tools, namely restriction fragment bias, restriction site heterozygosity and PCR GC content bias. We explore the performance of existing analysis tools given these biases and discuss approaches to limiting or handling biases in RAD-Seq data. While these biases need to be taken seriously, we believe RAD loci affected by them can be excluded or processed with relative ease in most cases and that most RAD loci will be accurately genotyped by existing tools.

Genome-wide RAD sequence data provide unprecedented resolution of species boundaries and relationships in the Lake Victoria cichlid adaptive radiation

Although population genomic studies using next generation sequencing (NGS) data are becoming increasingly common, studies focusing on phylogenetic inference using these data are in their infancy. Here, we use NGS data generated from reduced representation genomic libraries of restriction-site-associated DNA (RAD) markers to infer phylogenetic relationships among 16 species of cichlid fishes from a single rocky island community within Lake Victoria's cichlid adaptive radiation. Previous attempts at sequence-based phylogenetic analyses in Victoria cichlids have shown extensive sharing of genetic variation among species and no resolution of species or higher-level relationships. These patterns have generally been attributed to the very recent origin (<15,000 years) of the radiation, and ongoing hybridization between species. We show that as we increase the amount of sequence data used in phylogenetic analyses, we produce phylogenetic trees with unprecedented resolution for this group. In trees derived from our largest data supermatrices (3 to >5.8 million base pairs in width), species are reciprocally monophyletic with high bootstrap support, and the majority of internal branches on the tree have high support. Given the difficulty of the phylogenetic problem that the Lake Victoria cichlid adaptive radiation represents, these results are striking. The strict interpretation of the topologies we present here warrants caution because many questions remain about phylogenetic inference with very large genomic data set and because we can with the current analysis not distinguish between effects of shared ancestry and post-speciation gene flow. However, these results provide the first conclusive evidence for the monophyly of species in the Lake Victoria cichlid radiation and demonstrate the power that NGS data sets hold to resolve even the most difficult of phylogenetic challenges.

Parallel tagged amplicon sequencing reveals major lineages and phylogenetic structure in the North American tiger salamander (Ambystoma tigrinum) species complex

Modern analytical methods for population genetics and phylogenetics are expected to provide more accurate results when data from multiple genome-wide loci are analysed. We present the results of an initial application of parallel tagged sequencing (PTS) on a next-generation platform to sequence thousands of barcoded PCR amplicons generated from 95 nuclear loci and 93 individuals sampled across the range of the tiger salamander (Ambystoma tigrinum) species complex. To manage the bioinformatic processing of this large data set (344 330 reads), we developed a pipeline that sorts PTS data by barcode and locus, identifies high-quality variable nucleotides and yields phased haplotype sequences for each individual at each locus. Our sequencing and bioinformatic strategy resulted in a genome-wide data set with relatively low levels of missing data and a wide range of nucleotide variation. structure analyses of these data in a genotypic format resulted in strongly supported assignments for the majority of individuals into nine geographically defined genetic clusters. Species tree analyses of the most variable loci using a multi-species coalescent model resulted in strong support for most branches in the species tree; however, analyses including more than 50 loci produced parameter sampling trends that indicated a lack of convergence on the posterior distribution. Overall, these results demonstrate the potential for amplicon-based PTS to rapidly generate large-scale data for population genetic and phylogenetic-based research.

Meiotic maps of sockeye salmon derived from massively parallel DNA sequencing

Background

Meiotic maps are a key tool for comparative genomics and association mapping studies. Next-generation sequencing and genotyping by sequencing are speeding the processes of SNP discovery and the development of new genetic tools, including meiotic maps for numerous species. Currently there are limited genetic resources for sockeye salmon, Oncorhynchus nerka. We develop the first dense meiotic map for sockeye salmon using a combination of novel SNPs found in restriction site associated DNA (RAD tags) and SNPs available from existing expressed sequence tag (EST) based assays.

Results

We discovered and genotyped putative SNPs in 3,430 RAD tags. We removed paralogous sequence variants leaving 1,672 SNPs; these were combined with 53 EST-based SNP genotypes for linkage mapping. The map contained 29 male and female linkage groups, consistent with the haploid chromosome number expected for sockeye salmon. The female map contains 1,057 loci spanning 4,896 cM, and the male map contains 1,118 loci spanning 4,220 cM. Regions of conservation with rainbow trout and synteny between the RAD based rainbow trout map and the sockeye salmon map were established.

Conclusions

Using RAD sequencing and EST-based SNP assays we successfully generated the first high density linkage map for sockeye salmon.

Footprints in time: comparative quantitative trait loci mapping of the pitcher-plant mosquito, Wyeomyia smithii

Identifying regions of the genome contributing to phenotypic evolution often involves genetic mapping of quantitative traits. The focus then turns to identifying regions of 'major' effect, overlooking the observation that traits of ecological or evolutionary relevance usually involve many genes whose individual effects are small but whose cumulative effect is large. Herein, we use the power of fully interfertile natural populations of a single species of mosquito to develop three quantitative trait loci (QTL) maps: one between two post-glacially diverged populations and two between a more ancient and a post-glacial population. All demonstrate that photoperiodic response is genetically a highly complex trait. Furthermore, we show that marker regressions identify apparently 'non-significant' regions of the genome not identified by composite interval mapping, that the perception of the genetic basis of adaptive evolution is crucially dependent upon genetic background and that the genetic basis for adaptive evolution of photoperiodic response is highly variable within contemporary populations as well as between anciently diverged populations.

Highly incomplete taxa can rescue phylogenetic analyses from the negative impacts of limited taxon sampling

Background

Phylogenies are essential to many areas of biology, but phylogenetic methods may give incorrect estimates under some conditions. A potentially common scenario of this type is when few taxa are sampled and terminal branches for the sampled taxa are relatively long. However, the best solution in such cases (i.e., sampling more taxa versus more characters) has been highly controversial. A widespread assumption in this debate is that added taxa must be complete (no missing data) in order to save analyses from the negative impacts of limited taxon sampling. Here, we evaluate whether incomplete taxa can also rescue analyses under these conditions (empirically testing predictions from an earlier simulation study).

Methodology/Principal Findings

We utilize DNA sequence data from 16 vertebrate species with well-established phylogenetic relationships. In each replicate, we randomly sample 4 species, estimate their phylogeny (using Bayesian, likelihood, and parsimony methods), and then evaluate whether adding in the remaining 12 species (which have 50, 75, or 90% of their data replaced with missing data cells) can improve phylogenetic accuracy relative to analyzing the 4 complete taxa alone. We find that in those cases where sampling few taxa yields an incorrect estimate, adding taxa with 50% or 75% missing data can frequently (>75% of relevant replicates) rescue Bayesian and likelihood analyses, recovering accurate phylogenies for the original 4 taxa. Even taxa with 90% missing data can sometimes be beneficial.

Conclusions

We show that adding taxa that are highly incomplete can improve phylogenetic accuracy in cases where analyses are misled by limited taxon sampling. These surprising empirical results confirm those from simulations, and show that the benefits of adding taxa may be obtained with unexpectedly small amounts of data. These findings have important implications for the debate on sampling taxa versus characters, and for studies attempting to resolve difficult phylogenetic problems.

Double digest RADseq: an inexpensive method for de novo SNP discovery and genotyping in model and non-model species

The ability to efficiently and accurately determine genotypes is a keystone technology in modern genetics, crucial to studies ranging from clinical diagnostics, to genotype-phenotype association, to reconstruction of ancestry and the detection of selection. To date, high capacity, low cost genotyping has been largely achieved via “SNP chip” microarray-based platforms which require substantial prior knowledge of both genome sequence and variability, and once designed are suitable only for those targeted variable nucleotide sites. This method introduces substantial ascertainment bias and inherently precludes detection of rare or population-specific variants, a major source of information for both population history and genotype-phenotype association. Recent developments in reduced-representation genome sequencing experiments on massively parallel sequencers (commonly referred to as RAD-tag or RADseq) have brought direct sequencing to the problem of population genotyping, but increased cost and procedural and analytical complexity have limited their widespread adoption. Here, we describe a complete laboratory protocol, including a custom combinatorial indexing method, and accompanying software tools to facilitate genotyping across large numbers (hundreds or more) of individuals for a range of markers (hundreds to hundreds of thousands). Our method requires no prior genomic knowledge and achieves per-site and per-individual costs below that of current SNP chip technology, while requiring similar hands-on time investment, comparable amounts of input DNA, and downstream analysis times on the order of hours. Finally, we provide empirical results from the application of this method to both genotyping in a laboratory cross and in wild populations. Because of its flexibility, this modified RADseq approach promises to be applicable to a diversity of biological questions in a wide range of organisms.

2b-RAD: a simple and flexible method for genome-wide genotyping

We describe 2b-RAD, a streamlined restriction site-associated DNA (RAD) genotyping method based on sequencing the uniform fragments produced by type IIB restriction endonucleases. Well-studied accessions of Arabidopsis thaliana were genotyped to validate the method's accuracy and to demonstrate fine-tuning of marker density as needed. The simplicity of the 2b-RAD protocol makes it particularly suitable for high-throughput genotyping as required for linkage mapping and profiling genetic variation in natural populations.

Inferring phylogenies from RAD sequence data

Reduced-representation genome sequencing represents a new source of data for systematics, and its potential utility in interspecific phylogeny reconstruction has not yet been explored. One approach that seems especially promising is the use of inexpensive short-read technologies (e.g., Illumina, SOLiD) to sequence restriction-site associated DNA (RAD)--the regions of the genome that flank the recognition sites of restriction enzymes. In this study, we simulated the collection of RAD sequences from sequenced genomes of different taxa (Drosophila, mammals, and yeasts) and developed a proof-of-concept workflow to test whether informative data could be extracted and used to accurately reconstruct "known" phylogenies of species within each group. The workflow consists of three basic steps: first, sequences are clustered by similarity to estimate orthology; second, clusters are filtered by taxonomic coverage; and third, they are aligned and concatenated for "total evidence" phylogenetic analysis. We evaluated the performance of clustering and filtering parameters by comparing the resulting topologies with well-supported reference trees and we were able to identify conditions under which the reference tree was inferred with high support. For Drosophila, whole genome alignments allowed us to directly evaluate which parameters most consistently recovered orthologous sequences. For the parameter ranges explored, we recovered the best results at the low ends of sequence similarity and taxonomic representation of loci; these generated the largest supermatrices with the highest proportion of missing data. Applications of the method to mammals and yeasts were less successful, which we suggest may be due partly to their much deeper evolutionary divergence times compared to Drosophila (crown ages of approximately 100 and 300 versus 60 Mya, respectively). RAD sequences thus appear to hold promise for reconstructing phylogenetic relationships in younger clades in which sufficient numbers of orthologous restriction sites are retained across species.

Next-generation phylogeography: a targeted approach for multilocus sequencing of non-model organisms

The field of phylogeography has long since realized the need and utility of incorporating nuclear DNA (nDNA) sequences into analyses. However, the use of nDNA sequence data, at the population level, has been hindered by technical laboratory difficulty, sequencing costs, and problematic analytical methods dealing with genotypic sequence data, especially in non-model organisms. Here, we present a method utilizing the 454 GS-FLX Titanium pyrosequencing platform with the capacity to simultaneously sequence two species of sea star (Meridiastra calcar and Parvulastra exigua) at five different nDNA loci across 16 different populations of 20 individuals each per species. We compare results from 3 populations with traditional Sanger sequencing based methods, and demonstrate that this next-generation sequencing platform is more time and cost effective and more sensitive to rare variants than Sanger based sequencing. A crucial advantage is that the high coverage of clonally amplified sequences simplifies haplotype determination, even in highly polymorphic species. This targeted next-generation approach can greatly increase the use of nDNA sequence loci in phylogeographic and population genetic studies by mitigating many of the time, cost, and analytical issues associated with highly polymorphic, diploid sequence markers.

Next-generation hybridization and introgression

Hybridization has a major role in evolution-from the introgression of important phenotypic traits between species, to the creation of new species through hybrid speciation. Molecular studies of hybridization aim to understand the class of hybrids and the frequency of introgression, detect the signature of ancient hybridization, and understand the behaviour of introgressed loci in their new genomic background. This often involves a large investment in the design and application of molecular markers, leading to a compromise between the depth and breadth of genomic data. New techniques designed to assay a large sub-section of the genome, in association with next-generation sequencing (NGS) technologies, will allow genome-wide hybridization and introgression studies in organisms with no prior sequence data. These detailed genotypic data will unite the breadth of sampling of loci characteristic of population genetics with the depth of sequence information associated with molecular phylogenetics. In this review, we assess the theoretical and methodological constraints that limit our understanding of natural hybridization, and promote the use of NGS for detecting hybridization and introgression between non-model organisms. We also make recommendations for the ways in which emerging techniques, such as pooled barcoded amplicon sequencing and restriction site-associated DNA tags, should be used to overcome current limitations, and enhance our understanding of this evolutionary significant process.

Comprehensive evaluation of SNP identification with the Restriction Enzyme-based Reduced Representation Library (RRL) method

Background

Restriction Enzyme-based Reduced Representation Library (RRL) method represents a relatively feasible and flexible strategy used for Single Nucleotide Polymorphism (SNP) identification in different species. It has remarkable advantage of reducing the complexity of the genome by orders of magnitude. However, comprehensive evaluation for actual efficacy of SNP identification by this method is still unavailable.

Results

In order to evaluate the efficacy of Restriction Enzyme-based RRL method, we selected Tsp 45I enzyme which covers 266 Mb flanking region of the enzyme recognition site according to in silico simulation on human reference genome, then we sequenced YH RRL after Tsp 45I treatment and obtained reads of which 80.8% were mapped to target region with an 20-fold average coverage, about 96.8% of target region was covered by at least one read and 257 K SNPs were identified in the region using SOAPsnp software.

Compared with whole genome resequencing data, we observed false discovery rate (FDR) of 13.95% and false negative rate (FNR) of 25.90%. The concordance rate of homozygote loci was over 99.8%, but that of heterozygote were only 92.56%. Repeat sequences and bases quality were proved to have a great effect on the accuracy of SNP calling, SNPs in recognition sites contributed evidently to the high FNR and the low concordance rate of heterozygote. Our results indicated that repeat masking and high stringent filter criteria could significantly decrease both FDR and FNR.

Conclusions

This study demonstrates that Restriction Enzyme-based RRL method was effective for SNP identification. The results highlight the important role of bias and the method-derived defects represented in this method and emphasize the special attentions noteworthy.

Targeted enrichment strategies for next-generation plant biology

PREMISE OF THE STUDY

The dramatic advances offered by modern DNA sequencers continue to redefine the limits of what can be accomplished in comparative plant biology. Even with recent achievements, however, plant genomes present obstacles that can make it difficult to execute large-scale population and phylogenetic studies on next-generation sequencing platforms. Factors like large genome size, extensive variation in the proportion of organellar DNA in total DNA, polyploidy, and gene number/redundancy contribute to these challenges, and they demand flexible targeted enrichment strategies to achieve the desired goals.

METHODS

In this article, we summarize the many available targeted enrichment strategies that can be used to target partial-to-complete organellar genomes, as well as known and anonymous nuclear targets. These methods fall under four categories: PCR-based enrichment, hybridization-based enrichment, restriction enzyme-based enrichment, and enrichment of expressed gene sequences.

KEY RESULTS

Examples of plant-specific applications exist for nearly all methods described. While some methods are well established (e.g., transcriptome sequencing), other promising methods are in their infancy (hybridization enrichment). A direct comparison of methods shows that PCR-based enrichment may be a reasonable strategy for accessing small genomic targets (e.g., ≤50 kbp), but that hybridization and transcriptome sequencing scale more efficiently if larger targets are desired.

CONCLUSIONS

While the benefits of targeted sequencing are greatest in plants with large genomes, nearly all comparative projects can benefit from the improved throughput offered by targeted multiplex DNA sequencing, particularly as the amount of data produced from a single instrument approaches a trillion bases per run.

RAD in the realm of next-generation sequencing technologies

The first North American RAD Sequencing and Genomics Symposium, sponsored by Floragenex (http://www.floragenex.com/radmeeting/), took place in Portland, Oregon (USA) on 19 April 2011. This symposium was convened to promote and discuss the use of restriction-site-associated DNA (RAD) sequencing technologies. RAD sequencing is one of several strategies recently developed to increase the power of data generated via short-read sequencing technologies by reducing their complexity (Baird et al. 2008; Huang et al. 2009; Andolfatto et al. 2011; Elshire et al. 2011). RAD sequencing, as a form of genotyping by sequencing, has been effectively applied in genetic mapping and quantitative trait loci (QTL) analyses in a range of organisms including nonmodel, genetically highly heterogeneous organisms (Table 1; Baird et al. 2008; Baxter et al. 2011; Chutimanitsakun et al. 2011; Pfender et al. 2011). RAD sequencing has recently found applications in phylogeography (Emerson et al. 2010) and population genomics (Hohenlohe et al. 2010). Considering the diversity of talks presented during this meeting, more developments are to be expected in the very near future.

Nuclear and plastid haplotypes suggest rapid diploid and polyploid speciation in the N Hemisphere Achillea millefolium complex (Asteraceae)

Background

Species complexes or aggregates consist of a set of closely related species often of different ploidy levels, whose relationships are difficult to reconstruct. The N Hemisphere Achillea millefolium aggregate exhibits complex morphological and genetic variation and a broad ecological amplitude. To understand its evolutionary history, we study sequence variation at two nuclear genes and three plastid loci across the natural distribution of this species complex and compare the patterns of such variations to the species tree inferred earlier from AFLP data.

Results

Among the diploid species of A. millefolium agg., gene trees of the two nuclear loci, ncpGS and SBP, and the combined plastid fragments are incongruent with each other and with the AFLP tree likely due to incomplete lineage sorting or secondary introgression. In spite of the large distributional range, no isolation by distance is found. Furthermore, there is evidence for intragenic recombination in the ncpGS gene. An analysis using a probabilistic model for population demographic history indicates large ancestral effective population sizes and short intervals between speciation events. Such a scenario explains the incongruence of the gene trees and species tree we observe. The relationships are particularly complex in the polyploid members of A. millefolium agg.

Conclusions

The present study indicates that the diploid members of A. millefolium agg. share a large part of their molecular genetic variation. The findings of little lineage sorting and lack of isolation by distance is likely due to short intervals between speciation events and close proximity of ancestral populations. While previous AFLP data provide species trees congruent with earlier morphological classification and phylogeographic considerations, the present sequence data are not suited to recover the relationships of diploid species in A. millefolium agg. For the polyploid taxa many hybrid links and introgression from the diploids are suggested.

Population genomic analysis of model and nonmodel organisms using sequenced RAD tags

The evolutionary processes of mutation, migration, genetic drift, and natural selection shape patterns of genetic variation among individuals, populations, and species, and they can do so differentially across genomes. The field of population genomics provides a comprehensive genome-scale view of these processes, even beyond traditional model organisms. Until recently, genome-wide studies of genetic variation have been prohibitively expensive. However, next-generation sequencing (NGS) technologies are revolutionizing the field of population genomics, allowing for genetic analysis at scales not previously possible even in organisms for which few genomic resources presently exist. To speed this revolution in evolutionary genetics, we and colleagues developed Restriction site Associated DNA (RAD) sequencing, a method that uses Illumina NGS to simultaneously type and score tens to hundreds of thousands of single nucleotide polymorphism (SNP) markers in hundreds of individuals for minimal investment of resources. The core molecular protocol is described elsewhere in this volume, which can be modified to suit a diversity of evolutionary genetic questions. In this chapter, we outline the conceptual framework of population genomics, relate genomic patterns of variation to evolutionary processes, and discuss how RAD sequencing can be used to study population genomics. In addition, we discuss bioinformatic considerations that arise from unique aspects of NGS data as compared to traditional marker based approaches, and we outline some general analytical approaches for RAD-seq and similar data, including a computational pipeline that we developed called Stacks. This software can be used for the analysis of RAD-seq data in organisms with and without a reference genome. Nonetheless, the development of analytical tools remains in its infancy, and further work is needed to fully quantify sampling variance and biases in these data types. As data-gathering technology continues to advance, our ability to understand genomic evolution in natural populations will be limited more by conceptual and analytical weaknesses than by the amount of molecular data.

Molecular delineation of species in the coral holobiont

The coral holobiont is a complex assemblage of organisms spanning a diverse taxonomic range including a cnidarian host, as well as various dinoflagellate, prokaryotic and acellular symbionts. With the accumulating information on the molecular diversity of these groups, binomial species classification and a reassessment of species boundaries for the partners in the coral holobiont is a logical extension of this work and will help enhance the capacity for comparative research among studies. To aid in this endeavour, we review the current literature on species diversity for the three best studied partners of the coral holobiont (coral, Symbiodinium, prokaryotes) and provide suggestions for future work on systematics within these taxa. We advocate for an integrative approach to the delineation of species using both molecular genetics in combination with phenetic characters. We also suggest that an a priori set of criteria be developed for each taxonomic group as no one species concept or accompanying set of guidelines is appropriate for delineating all members of the coral holobiont.

SNP discovery in non-model organisms using 454 next generation sequencing

Roche 454 sequencing of the transcriptome has become a standard approach for efficiently obtaining single nucleotide polymorphisms (SNPs) in non-model species. In this chapter, the primary issues facing the development of SNPs from the transcriptome in non-model species are presented: tissue and sampling choices, mRNA preparation, considerations of normalization, pooling and barcoding, how much to sequence, how to assemble the data and assess the assembly, calling transcriptome SNPs, developing these into genomic SNPs, and publishing the work. Discussion also covers the comparison of this approach to RAD-tag sequencing and the potential of using other sequencing platforms for SNP development.

RAD paired-end sequencing for local de novo assembly and SNP discovery in non-model organisms

Restriction-site Associated DNA (RAD) markers are rapidly becoming a standard for SNP discovery and genotyping studies even in organisms without a sequenced reference genome. It is difficult, however, to identify genes nearby RAD markers of interest or move from SNPs identified by RAD to a high-throughput genotyping assay. Paired-end sequencing of RAD fragments can alleviate these problems by generating a set of paired sequences that can be locally assembled into high-quality contigs up to 1 kb in length. These contigs can then be used for SNP identification, homology searching, or high-throughput assay primer design. In this chapter, we offer suggestions on how to design a RAD paired-end (RAD-PE) sequencing project and the protocol for creating paired-end RAD libraries suitable for Illumina sequencers.

SNP discovery and genotyping for evolutionary genetics using RAD sequencing

Next-generation sequencing technologies are revolutionizing the field of evolutionary biology, opening the possibility for genetic analysis at scales not previously possible. Research in population genetics, quantitative trait mapping, comparative genomics, and phylogeography that was unthinkable even a few years ago is now possible. More importantly, these next-generation sequencing studies can be performed in organisms for which few genomic resources presently exist. To speed this revolution in evolutionary genetics, we have developed Restriction site Associated DNA (RAD) genotyping, a method that uses Illumina next-generation sequencing to simultaneously discover and score tens to hundreds of thousands of single-nucleotide polymorphism (SNP) markers in hundreds of individuals for minimal investment of resources. In this chapter, we describe the core RAD-seq protocol, which can be modified to suit a diversity of evolutionary genetic questions. In addition, we discuss bioinformatic considerations that arise from unique aspects of next-generation sequencing data as compared to traditional marker-based approaches, and we outline some general analytical approaches for RAD-seq and similar data. Despite considerable progress, the development of analytical tools remains in its infancy, and further work is needed to fully quantify sampling variance and biases in these data types.

Next generation quantitative genetics in plants

Most characteristics in living organisms show continuous variation, which suggests that they are controlled by multiple genes. Quantitative trait loci (QTL) analysis can identify the genes underlying continuous traits by establishing associations between genetic markers and observed phenotypic variation in a segregating population. The new high-throughput sequencing (HTS) technologies greatly facilitate QTL analysis by providing genetic markers at genome-wide resolution in any species without previous knowledge of its genome. In addition HTS serves to quantify molecular phenotypes, which aids to identify the loci responsible for QTLs and to understand the mechanisms underlying diversity. The constant improvements in price, experimental protocols, computational pipelines, and statistical frameworks are making feasible the use of HTS for any research group interested in quantitative genetics. In this review I discuss the application of HTS for molecular marker discovery, population genotyping, and expression profiling in QTL analysis.

Genomic diversity in switchgrass (Panicum virgatum): from the continental scale to a dune landscape

Connecting broad-scale patterns of genetic variation and population structure to genetic diversity on a landscape is a key step towards understanding historical processes of migration and adaptation. New genomic approaches can be used to increase the resolution of phylogeographic studies while reducing locus sampling effects and circumventing ascertainment bias. Here, we use a novel approach based on high-throughput sequencing to characterize genetic diversity in complete chloroplast genomes and >10,000 nuclear loci in switchgrass, at continental and landscape scales. Switchgrass is a North American tallgrass species, which is widely used in conservation and perennial biomass production, and shows strong ecotypic adaptation and population structure across the continental range. We sequenced 40.9 billion base pairs from 24 individuals from across the species' range and 20 individuals from the Indiana Dunes. Analysis of plastome sequence revealed 203 variable SNP sites that define eight haplogroups, which are differentiated by 4-127 SNPs and confirmed by patterns of indel variation. These include three deeply divergent haplogroups, which correspond to the previously described lowland-upland ecotypic split and a novel upland haplogroup split that dates to the mid-Pleistocene. Most of the plastome haplogroup diversity present in the northern switchgrass range, including in the Indiana Dunes, originated in the mid- or upper Pleistocene prior to the most recent postglacial recolonization. Furthermore, a recently colonized landscape feature (approximately 150 ya) in the Indiana Dunes contains several deeply divergent upland haplogroups. Nuclear markers also support a deep lowland-upland split, followed by limited gene flow, and show extensive gene flow in the local population of the Indiana Dunes.

The promiscuous and the chaste: frequent allopolyploid speciation and its genomic consequences in American daisies (Melampodium sect. Melampodium; Asteraceae)

Polyploidy, an important factor in eukaryotic evolution, is especially abundant in angiosperms, where it often acts in concert with hybridization to produce allopolyploids. The application of molecular phylogenetic techniques has identified the origins of numerous allopolyploids, but little is known on genomic and chromosomal consequences of allopolyploidization, despite their important role in conferring divergence of allopolyploids from their parental species. Here, using several plastid and nuclear sequence markers, we clarify the origin of tetra- and hexaploids in a group of American daisies, allowing characterization of genome dynamics in polyploids compared to their diploid ancestors. All polyploid species are allopolyploids. Among the four diploid gene pools, the propensity for allopolyploidization is unevenly distributed phylogenetically with a few species apparently more prone to participate, but the underlying causes remain unclear. Polyploid genomes are characterized by differential loss of ribosomal DNA loci (5S and 35S rDNA), known hotspots of chromosomal evolution, but show genome size additivity, suggesting limited changes beyond those affecting rDNA loci or the presence of processes counterbalancing genome reduction. Patterns of rDNA sequence conversion and provenance of the lost loci are highly idiosyncratic and differ even between allopolyploids of identical parentage, indicating that allopolyploids deriving from the same lower-ploid parental species can follow different evolutionary trajectories.

Stacks: building and genotyping Loci de novo from short-read sequences

Advances in sequencing technology provide special opportunities for genotyping individuals with speed and thrift, but the lack of software to automate the calling of tens of thousands of genotypes over hundreds of individuals has hindered progress. Stacks is a software system that uses short-read sequence data to identify and genotype loci in a set of individuals either de novo or by comparison to a reference genome. From reduced representation Illumina sequence data, such as RAD-tags, Stacks can recover thousands of single nucleotide polymorphism (SNP) markers useful for the genetic analysis of crosses or populations. Stacks can generate markers for ultra-dense genetic linkage maps, facilitate the examination of population phylogeography, and help in reference genome assembly. We report here the algorithms implemented in Stacks and demonstrate their efficacy by constructing loci from simulated RAD-tags taken from the stickleback reference genome and by recapitulating and improving a genetic map of the zebrafish, Danio rerio.

Dynamic evolution of base composition: causes and consequences in avian phylogenomics

Resolving the phylogenetic relationships among birds is a classical problem in systematics, and this is particularly so when it comes to understanding the relationships among Neoaves. Previous phylogenetic inference of birds has been limited to mitochondrial genomes or a few nuclear genes. Here, we apply deep brain transcriptome sequencing of nine bird species (several passerines, hummingbirds, dove, parrot, and emu), using next-generation sequencing technology to understand features of transcriptome evolution in birds and how this affects phylogenetic inference, and combine with data from two bird species using first generation technology. The phylogenomic data matrix comprises 1,995 genes and a total of 0.77 Mb of exonic sequence. First, we find an unexpected heterogeneity in the evolution of base composition among avian lineages. There is a pronounced increase in guanine + cytosine (GC) content in the third codon position in several independent lineages, with the strongest effect seen in passerines. Second, we evaluate the effect of GC content variation on phylogenetic reconstruction. We find important inconsistencies between the topologies obtained with or without taking GC variation into account, each supporting different conclusions of past studies and also influencing hypotheses on the evolution of the trait of vocal learning. Third, we demonstrate a link between GC content evolution and recombination rate and, focusing on the zebra finch lineage, find that recombination seems to drive GC content. Although we cannot reveal the causal relationships, this observation is consistent with the model of GC-biased gene conversion. Finally, we use this unparalleled amount of avian sequence data to study the rate of molecular evolution, calibrated by fossil evidence and augmented with data from alligator transcriptome sequencing. There is a 2- to 3-fold variation in substitution rate among lineages with passerines being the most rapidly evolving and ratites the slowest. This study illustrates the potential of next-generation sequencing for phylogenomic studies but also the pitfalls when using genome-wide data with heterogeneous base composition.

A re-appraisal of nunatak survival in arctic-alpine phylogeography

A long standing and at times fervid debate in biogeography revolves around the question whether arctic and high alpine organisms survived Pleistocene ice ages on small island-like areas protruding above the ice-sheet, socalled nunataks, or whether they did so in peripheral nonglaciated refugial areas. A common picture emerging from a plethora of molecular phylogeographic studies in the last decade is that both in the Arctic and in temperate mountain ranges such as the European Alps nunatak survival needs to be only rarely invoked to explain observed genetic patterns (for a rare example see Stehlik et al. 2002). As two studies in this issue show, depreciation of the nunatak hypothesis is, however, not warranted. In this issue of Molecular Ecology Westergaard et al. (2011) investigate genetic patterns of two arctic-alpine plant species distributed on both sides of the Atlantic exclusively in areas that were mostly covered by ice-sheets during Pleistocene glacial advances. In both species, amplified fragment length polymorphism (AFLP) data identified divergent and partly genetically diverse groups east and west of the Atlantic. This suggests, for the first time in Arctic plants, in situ survival on nunataks. In an entirely different geographic setting and on a different geographic scale, Lohse et al. (2011, this issue) study the colonization of high alpine areas in the Orobian Alps, situated within and adjacent to a prominent peripheral refugial area (massif de refuge) in the Southern Alps of northern Italy, by dispersal-limited carabid ground beetles. Using explicit hypothesis testing and inference of ancestral locations in a Bayesian framework, stepwise colonization from two separate southern refugia is found to shape the genetic pattern of these beetles, but at the northern edge, populations survived at least parts of the last glaciation in situ on nunataks.

Paired-end RAD-seq for de novo assembly and marker design without available reference

MOTIVATION

Next-generation sequencing technologies have facilitated the study of organisms on a genome-wide scale. A recent method called restriction site associated DNA sequencing (RAD-seq) allows to sample sequence information at reduced complexity across a target genome using the Illumina platform. Single-end RAD-seq has proven to provide a large number of informative genetic markers in reference as well as non-reference organisms.

RESULTS

Here, we present a method for de novo assembly of paired-end RAD-seq data in order to produce extended contigs flanking a restriction site. We were able to reconstruct one-tenth of the guppy genome represented by 200-500 bp contigs associated to EcoRI recognition sites. In addition, these contigs were used as reference allowing the detection of thousands of new polymorphic markers that are informative for mapping and population genetic studies in the guppy.

AVAILABILITY

A perl and C++ implementation of the method demonstrated in this article is available under http://guppy.weigelworld.org/weigeldatabases/radMarkers/ as package RApiD.

CONTACT

christine.dreyer@tuebingen.mpg.de

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Genome-wide genetic marker discovery and genotyping using next-generation sequencing

The advent of next-generation sequencing (NGS) has revolutionized genomic and transcriptomic approaches to biology. These new sequencing tools are also valuable for the discovery, validation and assessment of genetic markers in populations. Here we review and discuss best practices for several NGS methods for genome-wide genetic marker development and genotyping that use restriction enzyme digestion of target genomes to reduce the complexity of the target. These new methods -- which include reduced-representation sequencing using reduced-representation libraries (RRLs) or complexity reduction of polymorphic sequences (CRoPS), restriction-site-associated DNA sequencing (RAD-seq) and low coverage genotyping -- are applicable to both model organisms with high-quality reference genome sequences and, excitingly, to non-model species with no existing genomic data.

Linkage mapping and comparative genomics using next-generation RAD sequencing of a non-model organism

Restriction-site associated DNA (RAD) sequencing is a powerful new method for targeted sequencing across the genomes of many individuals. This approach has broad potential for genetic analysis of non-model organisms including genotype-phenotype association mapping, phylogeography, population genetics and scaffolding genome assemblies through linkage mapping. We constructed a RAD library using genomic DNA from a Plutella xylostella (diamondback moth) backcross that segregated for resistance to the insecticide spinosad. Sequencing of 24 individuals was performed on a single Illumina GAIIx lane (51 base paired-end reads). Taking advantage of the lack of crossing over in homologous chromosomes in female Lepidoptera, 3,177 maternally inherited RAD alleles were assigned to the 31 chromosomes, enabling identification of the spinosad resistance and W/Z sex chromosomes. Paired-end reads for each RAD allele were assembled into contigs and compared to the genome of Bombyx mori (n = 28) using BLAST, revealing 28 homologous matches plus 3 expected fusion/breakage events which account for the difference in chromosome number. A genome-wide linkage map (1292 cM) was inferred with 2,878 segregating RAD alleles inherited from the backcross father, producing chromosome and location specific sequenced RAD markers. Here we have used RAD sequencing to construct a genetic linkage map de novo for an organism that has no previous genome data. Comparative analysis of P. xyloxtella linkage groups with B. mori chromosomes shows for the first time, genetic synteny appears common beyond the Macrolepidoptera. RAD sequencing is a powerful system capable of rapidly generating chromosome specific data for non-model organisms.

RADSeq: next-generation population genetics

Next-generation sequencing technologies are making a substantial impact on many areas of biology, including the analysis of genetic diversity in populations. However, genome-scale population genetic studies have been accessible only to well-funded model systems. Restriction-site associated DNA sequencing, a method that samples at reduced complexity across target genomes, promises to deliver high resolution population genomic data-thousands of sequenced markers across many individuals-for any organism at reasonable costs. It has found application in wild populations and non-traditional study species, and promises to become an important technology for ecological population genomics.

Local de novo assembly of RAD paired-end contigs using short sequencing reads

Despite the power of massively parallel sequencing platforms, a drawback is the short length of the sequence reads produced. We demonstrate that short reads can be locally assembled into longer contigs using paired-end sequencing of restriction-site associated DNA (RAD-PE) fragments. We use this RAD-PE contig approach to identify single nucleotide polymorphisms (SNPs) and determine haplotype structure in threespine stickleback and to sequence E. coli and stickleback genomic DNA with overlapping contigs of several hundred nucleotides. We also demonstrate that adding a circularization step allows the local assembly of contigs up to 5 kilobases (kb) in length. The ease of assembly and accuracy of the individual contigs produced from each RAD site sequence suggests RAD-PE sequencing is a useful way to convert genome-wide short reads into individually-assembled sequences hundreds or thousands of nucleotides long.