Sarah M. Hird

Applications of next-generation sequencing to phylogeography and phylogenetics

This is a time of unprecedented transition in DNA sequencing technologies. Next-generation sequencing (NGS) clearly holds promise for fast and cost-effective generation of multilocus sequence data for phylogeography and phylogenetics. However, the focus on non-model organisms, in addition to uncertainty about which sample preparation methods and analyses are appropriate for different research questions and evolutionary timescales, have contributed to a lag in the application of NGS to these fields. Here, we outline some of the major obstacles specific to the application of NGS to phylogeography and phylogenetics, including the focus on non-model organisms, the necessity of obtaining orthologous loci in a cost-effective manner, and the predominate use of gene trees in these fields. We describe the most promising methods of sample preparation that address these challenges. Methods that reduce the genome by restriction digest and manual size selection are most appropriate for studies at the intraspecific level, whereas methods that target specific genomic regions (i.e., target enrichment or sequence capture) have wider applicability from the population level to deep-level phylogenomics. Additionally, we give an overview of how to analyze NGS data to arrive at data sets applicable to the standard toolkit of phylogeography and phylogenetics, including initial data processing to alignment and genotype calling (both SNPs and loci involving many SNPs). Even though whole-genome sequencing is likely to become affordable rather soon, because phylogeography and phylogenetics rely on analysis of hundreds of individuals in many cases, methods that reduce the genome to a subset of loci should remain more cost-effective for some time to come.

Ancient hybridization and mitochondrial capture between two species of chipmunks

Models that posit speciation in the face of gene flow are replacing classical views that hybridization is rare between animal species. We use a multilocus approach to examine the history of hybridization and gene flow between two species of chipmunks (Tamias ruficaudus and T. amoenus). Previous studies have shown that these species occupy different ecological niches and have distinct genital bone morphologies, yet appear to be incompletely isolated reproductively in multiple areas of sympatry. We compared data from four sequenced nuclear loci and from seven microsatellite loci to published cytochrome b sequences. Interspecific gene flow was primarily restricted to introgression of the T. ruficaudus mitochondrial genome into a sympatric subspecies of T. amoenus, T. a. canicaudus, with the four sequenced nuclear loci showing little to no interspecific allele sharing. Microsatellite data were consistent with high levels of differentiation between the species and also showed no current gene flow between broadly sympatric populations of T. a. canicaudus and T. ruficaudus. Coalescent analyses date the mtDNA introgression event from the mid‐Pleistocene to late Pliocene. Overall, these data indicate that introgression has had a minimal impact on the nuclear genomes of T. amoenus and T. ruficaudus despite multiple independent hybridization events. Our findings challenge long‐standing assumptions on patterns of reproductive isolation in chipmunks and suggest that there may be other examples of hybridization among the 23 species of Tamias that occur in western North America.

Next-generation sequencing reveals phylogeographic structure and a species tree for recent bird divergences

Next generation sequencing (NGS) technologies are revolutionizing many biological disciplines but have been slow to take root in phylogeography. This is partly due to the difficulty of using NGS to sequence orthologous DNA fragments for many individuals at low cost. We explore cases of recent divergence in four phylogenetically diverse avian systems using a method for quick and cost-effective generation of primary DNA sequence data using pyrosequencing. NGS data were processed using an analytical pipeline that reduces many reads into two called alleles per locus per individual. Using single nucleotide polymorphisms (SNPs) mined from the loci, we detected population differentiation in each of the four bird systems, including: a case of ecological speciation in rails (Rallus); a rapid postglacial radiation in the genus Junco; recent in situ speciation among hummingbirds (Trochilus) in Jamaica; and subspecies of white-crowned sparrows (Zonotrichia leucophrys) along the Pacific coast. The number of recovered loci aligning closely to chromosomal locations on the zebra finch (Taeniopygia guttata) genome was highly correlated to the size of the chromosome, suggesting that loci are randomly distributed throughout the genome. Using eight loci found in Zonotrichia and Junco lineages, we were also able to generate a species tree of these sparrow sister genera, demonstrating the potential of this method for generating data amenable to coalescent-based analysis. We discuss improvements that should enhance the methods utility for primary data generation.

The Carnivorous Pale Pitcher Plant Harbors Diverse, Distinct, and Time-Dependent Bacterial Communities

ABSTRACT The ability of American carnivorous pitcher plants (Sarracenia) to digest insect prey is facilitated by microbial associations. Knowledge of the details surrounding this interaction has been limited by our capability to characterize bacterial diversity in this system. To describe microbial diversity within and between pitchers of one species, Sarracenia alata, and to explore how these communities change over time as pitchers accumulate and digest insect prey, we collected and analyzed environmental sequence tag (454 pyrosequencing) and genomic fingerprint (automated ribosomal intergenic spacer analysis and terminal restriction fragment length polymorphism) data. Microbial richness associated with pitcher plant fluid is high; more than 1,000 unique phylogroups were identified across at least seven phyla and 50 families. We documented an increase in bacterial diversity and abundance with time and observed repeated changes in bacterial community composition. Pitchers from different plants harbored significantly more similar bacterial communities at a given time point than communities coming from the same genetic host over time. The microbial communities in pitcher plant fluid also differ significantly from those present in the surrounding soil. These findings indicate that the bacteria associated with pitcher plant leaves are far from random assemblages and represent an important step toward understanding this unique plant-microbe interaction.

Poor Fit to the Multispecies Coalescent is Widely Detectable in Empirical Data

Model checking is a critical part of Bayesian data analysis, yet it remains largely unused in systematic studies. Phylogeny estimation has recently moved into an era of increasingly complex models that simultaneously account for multiple evolutionary processes, the statistical fit of these models to the data has rarely been tested. Here we develop a posterior predictive simulation-based model check for a commonly used multispecies coalescent model, implemented in *BEAST, and apply it to 25 published data sets. We show that poor model fit is detectable in the majority of data sets; that this poor fit can mislead phylogenetic estimation; and that in some cases it stems from processes of inherent interest to systematists. We suggest that as systematists scale up to phylogenomic data sets, which will be subject to a heterogeneous array of evolutionary processes, critically evaluating the fit of models to data is an analytical step that can no longer be ignored.

Deep Phylogeographic Structure and Environmental Differentiation in the Carnivorous Plant Sarracenia alata

We collected ~29 kb of sequence data using Roche 454 pyrosequencing in order to estimate the timing and pattern of diversification in the carnivorous pitcher plant Sarracenia alata. Utilizing modified protocols for reduced representation library construction, we generated sequence data from 86 individuals across 10 populations from throughout the range of the species. We identified 76 high-quality and high-coverage loci (containing over 500 SNPs) using the bioinformatics pipeline PRGmatic. Results from a Bayesian clustering analysis indicate that populations are highly structured, and are similar in pattern to the topology of a population tree estimated using *BEAST. The pattern of diversification within Sarracenia alata implies that riverine barriers are the primary factor promoting population diversification, with divergence across the Mississippi River occurring more than 60,000 generations before present. Further, significant patterns of niche divergence and the identification of several outlier loci suggest that selection may contribute to population divergence. Our results demonstrate the feasibility of using next-generation sequencing to investigate intraspecific genetic variation in nonmodel species.

Comparative Gut Microbiota of 59 Neotropical Bird Species

The gut microbiota of vertebrates are essential to host health. Most non-model vertebrates, however, lack even a basic description of natural gut microbiota biodiversity. Here, we sampled 116 intestines from 59 Neotropical bird species and used the V6 region of the 16S rRNA molecule as a microbial fingerprint (average coverage per bird ~80,000 reads). A core microbiota of Proteobacteria, Firmicutes, Bacteroidetes, and Actinobacteria was identified, as well as several gut-associated genera. We tested 18 categorical variables associated with each bird for significant correlation to the gut microbiota; host taxonomic categories were most frequently significant and explained the most variation. Ecological variables (e.g., diet, foraging stratum) were also frequently significant but explained less variation. Little evidence was found for a significant influence of geographic space. Finally, we suggest that microbial sampling during field collection of organisms would propel biological understanding of evolutionary history and ecological significance of host-associated microbiota.

Sampling locality is more detectable than taxonomy or ecology in the gut microbiota of the brood-parasitic Brown-headed Cowbird (Molothrus ater)

Brown-headed Cowbirds (Molothrus ater) are the most widespread avian brood parasite in North America, laying their eggs in the nests of approximately 250 host species that raise the cowbird nestlings as their own. It is currently unknown how these heterospecific hosts influence the cowbird gut microbiota relative to other factors, such as the local environment and genetics. We test a Nature Hypothesis (positing the importance of cowbird genetics) and a Nurture Hypothesis (where the host parents are most influential to cowbird gut microbiota) using the V6 region of 16S rRNA as a microbial fingerprint of the gut from 32 cowbird samples and 16 potential hosts from nine species. We test additional hypotheses regarding the influence of the local environment and age of the birds. We found no evidence for the Nature Hypothesis and little support for the Nurture Hypothesis. Cowbird gut microbiota did not form a clade, but neither did members of the host species. Rather, the physical location, diet and age of the bird, whether cowbird or host, were the most significant categorical variables. Thus, passerine gut microbiota may be most strongly influenced by environmental factors. To put this variation in a broader context, we compared the bird data to a fecal microbiota dataset of 38 mammal species and 22 insect species. Insects were always the most variable; on some axes, we found more variation within cowbirds than across all mammals. Taken together, passerine gut microbiota may be more variable and environmentally determined than other taxonomic groups examined to date.

Assessment of gene flow across a hybrid zone in red-tailed chipmunks (Tamias ruficaudus).

The role of hybridization in animal speciation is controversial and recent research has challenged the long‐standing criterion of complete reproductive isolation to attain species status. The speciation‐with‐gene‐flow model posits that the genome is semi‐permeable and hybridization may be a phase in the process of divergence. Here, we apply these concepts to a previously identified zone of mtDNA introgression between the two strongly morphologically differentiated subspecies of red‐tailed chipmunk (Tamias ruficaudus) in the US Inland Northwest. Using multilocus genotype data from the southern, older contact zone, we demonstrate that neutral gene flow is unusually low between the subspecies across the Lochsa River. This is geographically congruent with the discontinuity in bacular morphology, indicating that the cline of mitochondrial DNA (mtDNA) haplotypes is displaced. Furthermore, we elucidate the evolutionary forces responsible by testing hypotheses of lineage sorting and hybridization. We determined that introgressive hybridization is the cause of mtDNA/morphology incongruence because there are non‐zero levels of migration and gene flow. Although our estimate of the age of the hybrid zone has wide credibility intervals, the hybridization events occurred in the Late Pleistocene and the divergence occurred in the Middle Pleistocene. Finally, we assessed substructure within and adjacent to the hybrid zone and found that the hybrid zone constitutes a set of populations that are genetically differentiated from parental sets of populations; therefore, hybridization in this system is not likely an evolutionary sink, but has generated novel combinations of genotypes.

PRGmatic: an efficient pipeline for collating genome-enriched second-generation sequencing data using a ‘provisional-reference genome’

Second‐generation sequencing is increasingly being used in combination with genome‐enrichment techniques to amplify a large number of loci in many individuals for the purpose of population genetic and phylogeographic analysis. Compiling all the necessary tools to analyse these data is complex and time‐consuming. Here, we assemble a set of programs and pipe them together with Perl, enabling research laboratories without a dedicated bioinformatician to utilize second‐generation sequencing. User input is a folder of the second‐generation sequencing reads sorted by individual (in FASTA format) and pipeline output is a folder of multi‐FASTA files that correspond to loci (with 2 alleles called per individual). Additional output includes a summary file of the number of individuals per locus, observed and expected heterozygosity for each locus, distribution of multiple hits and summary statistics (θ, Tajima’s D, etc.). This user‐friendly, open source pipeline, which requires no a priori reference genome because it constructs its own, allows the user to set various parameters (e.g. minimum coverage) in the dependent programs (CAP3, BWA, SAMtools and VarScan) and facilitates evaluation of the nature and quality of data collected prior to analysis in software packages.