David P. L. Toews

A bstract: Understanding the mechanisms facilitating the repeated evolution of traits is a core question in evolutionary biology. While repeated evolution has been observed across many systems, behavioral phenotypes have been less well studied than morphological traits. Migration is a behavioral adaptation to seasonality that has evolved independently in many animals, and repeated gains and losses of migratory behavior have been especially prevalent throughout the diversification of birds. While migratory traits are heritable in many taxa and candidate genes underlying intraspecific variation in migratory behavior have been identified, whether similar genetic changes are responsible for the repeated evolution of migration remains poorly understood. We used a phylogenomic approach to investigate the relationship between variation in candidate genes and migratory propensity across nearly all species in the parulid warbler radiation ( n = 101), a group of songbirds in which migratory behavior has been lost and gained multiple times. Comparisons of candidate gene tree topologies to the parulid warbler species tree revealed cases of discordance, though these were explained by differences in the number of phylogenetically informative sites among genes rather than by elevated genetic similarity among species sharing migratory strategies. Testing models of coding sequence evolution identified sites with elevated dN/dS in five candidate genes, suggestive of positive selection; however, variation at these sites showed no clear association with migratory strategy, nor did we detect recurrent selection across lineages where transitions in migratory behavior most likely occurred. Together, these results contradict the hypothesis that selection on shared variation (ancestral polymorphisms or introgressed alleles) facilitated the repeated evolution of migration in this clade.

The gut microbiome is a dynamic ecosystem wherein microbes can exert beneficial, neutral or harmful effects on their host organism. Previous research has supported a large role for the environment in shaping avian gut microbiome diversity, but host‐specific factors that regulate gut microbiome variation remain elusive. In hybrid zones, genetic recombination shuffles divergent alleles among backcrossed individuals allowing associations between genomic regions and specific traits to be identified. In this study, we use an association mapping approach to investigate the contribution of host alleles in shaping gut microbiome composition. We collected samples from across the Yellow‐rumped Warbler hybrid zone ( Setophaga coronata coronata × S. c. auduboni ), including from distantly allopatric sites. The narrow width of this hybrid zone suggests that selection acts against hybrids, although the source of that selection is unclear. We quantified gut microbiome variation using 16S amplicon sequencing and produced genome‐wide sequence data for hosts to link warbler genotypes to microbiome traits. This study is one of the first to identify candidate genes underlying gut microbiome variation in wild passerines. Notably, candidate loci include genes with immune function, redox status and gene regulation functions; two genes overlap with candidate genes identified in another avian system. Genetic differentiation was weak among candidate loci, indicating that alleles associated with gut microbiome variation are shared between subspecies. Our analysis of microbiome variation across nearly the full breeding range of an avian species complex yields important insights on the genetic factors that shape symbiotic interactions in vertebrate systems.

Assessing genetic differentiation and its drivers among populations is essential for informing localized conservation management. Here, we study Olrog’s gull (Larus atlanticus), a threatened seabird that breeds in two sectors of coastal Argentina separated by approximately 700 km (Buenos Aires and Chubut provinces). We used a reduced-representation genomic approach to genotype 59 individuals from five colonies at thousands of single-nucleotide polymorphisms to study patterns of connectivity across the species range. While individuals from the four northern colonies are genetically indistinguishable, the single colony located in the Patagonian province of Chubut is highly genetically differentiated. These individuals also have low genetic diversity, high inbreeding, and are closely related (higher than first-order relatives). For a subset of individuals, we generated whole genome sequences to assess the possible effects of inbreeding on the depletion of genetic variation across the genome (i.e. the number and length of runs of homozygosity (ROH)). We found both a higher number and longer ROH in the Chubut colony compared to its northern counterparts. This southern colony is genetically isolated from the panmictic northern colonies containing most of the breeding population of Orlog’s gull, and at risk of suffering from inbreeding depression and therefore of significant conservation concern.

Strong population contractions can leave a persistent genomic legacy that can influence populations long after their demographic recovery. While bottlenecks facilitate the removal of strongly deleterious mutations, the effectiveness of purging may be limited in historically small populations. The Kirtland’s warbler ( Setophaga kirtlandii ) is a rare North American songbird with an ancestrally small population. After narrowly evading extinction, they are one of few species that have been delisted from federal protections in the USA. Despite their recovery, a previous study showed evidence for recent inbreeding and a high burden of deleterious mutations that may have not been purged despite strong bottlenecks. Historical DNA offers a unique opportunity to understand the consequences of recent demographic declines on genetic diversity. Here, we use DNA from over 100-year-old museum specimens to estimate changes in genetic load in the Kirtland’s warblers pre- and post-bottleneck. We validate our results with forward-in-time genetic simulations and explore how sample size and missing data can affect estimates. Both empirical data and simulations suggest a reduced ability to purge deleterious mutations in this historically small population. Our simulations also highlight that limited sampling design and data quality can constrain the ability to detect changes.

Introgression between species has the potential to shape evolutionary trajectories in important ways, but uncovering complex introgression dynamics has only recently been made possible by advances in genomics. Warblers of the avian family Parulidae exemplify rapid diversification and sexual trait divergence, and we endeavored to study historical introgression in the family. We sequenced multiple genomes of nearly every species, constructed a phylogeny for the family, and investigated gene flow across the genome and at genes known for controlling feather color. We found that DNA haplotypes including the gene BCO2 , which encodes an enzyme that degrades yellow carotenoids, have spread among genera multiple times—from Vermivora to Geothlypis and from Leiothlypis to multiple Cardellina and Setophaga species. Patterns of inheritance in the latter case point to introgression that occurred 0.5 to 2 million years ago, and the shared haplotype among recipient species is less than 100 nucleotides long. Separately, we found evidence of introgression from red Cardellina species to both of the two red Myioborus species at BDH1L and from one red Myioborus species to the other at CYP2J19 ; both are key genes in the pathway that converts yellow carotenoids to red ketocarotenoids. Our results show that introgression is a common mechanism for the evolution of colorful plumage in parulid warblers and hint at complex histories of gene flow behind some of the Western Hemisphere’s most colorful birds.

The role of diet in driving ecological differences across the radiation of parulid wood warblers has been a topic of substantial debate. The inferences made in Robert MacArthur’s original study of their niche differences relied heavily on observations from microhabitat partitioning. How these different foraging behaviors translated to distinct diets and, more specifically, individual prey items has been less clear. Here, we use fecal metabarcoding of 2 syntopic insectivorous warblers—Setophaga virens (black-throated Green Warbler) and Seiurus aurocapilla (Ovenbird)—to complement and expand previous work, and we address past limitations by densely sampling many birds at the same location over a short period of time. We found highly significant differences in diet composition using several multivariate measures of diversity. In an analysis of individual diet proportions of insect orders, S. aurocapilla consumed more beetles and flies (Coleoptera and Diptera), whereas S. virens consumed more “true bugs” (Hemiptera) and stoneflies (Plecoptera). At the arthropod species level, we found that both warblers readily consumed invasive spongy moths (Lymantria dispar), and we identified 9 other arthropod species that significantly differed between the warblers. Of those, for 3 spider taxa, we combined warbler diet information with observations from arthropod collections and showed that spiders, which were more likely to be encountered on the ground, were exclusively eaten by S. aurocapilla whereas those encountered in the canopy were more likely to be consumed by S. virens, fitting with the expected vertical foraging stratification of the warblers. We interpret these diet differences as likely due to these 2 warbler species “opportunistically” encountering different arthropod assemblages in distinct foraging strata as opposed to “preferentially” consuming different prey. Our research emphasizes the benefits of extending analyses to more distantly related taxa—beyond those considered by MacArthur—and suggests a need for similar fine-scale studies within genera to enhance our understanding of dietary dynamics.

Owing in large part to Robert MacArthur’s classic research, wood warblers in the family Parulidae are textbook exemplars of species competition and niche partitioning. Conventional wisdom suggests that subtle differences in foraging behaviour are the principal means by which these nearly morphologically indistinguishable species are able to co-occur and avoid extinction. Yet, MacArthur’s study was in fact quite limited in scale, and he said little about the relevance of evolution to the study system. Here, we reassess MacArthur’s conclusions across an expanded set of syntopic warbler species in a forest in northern New York. We combine morphometrics, quantitative foraging data and faecal metabarcoding—a direct measure of warbler diet—to study competition and niche partitioning in an evolutionary framework. We find close relationships between morphology and foraging behaviour, but little connection between warbler ecomorphology and the 2237 invertebrate taxa detected in their diets. Instead, diet remains phylogenetically conserved—closely related warblers eat similar suites of invertebrates, regardless of where they forage. Finally, we present evidence that these species not only partition niche space in the present day but also that competition has shaped their behaviours over evolutionary time.

The dynamics of hybridization between golden-winged ( Vermivora chrysoptera ) and blue-winged warblers ( V. cyanoptera ) has been of interest for over a century. Whole genome analysis found only a small number of genomic regions that differed between the species. We previously developed a restriction enzyme-based RFLP approach to genotype large numbers of individuals at each of these loci. Here we extend this approach to an amplicon sequencing method to genotype individuals at these six plumage-associated regions. We demonstrate the efficacy using preliminary data from 4 golden-winged and 4 blue-winged warblers as well as provide the data and scripts necessary to analyze these data for other interested in replicating this approach. Our hope is that these data are useful for other researchers interested in genotyping Vermivora warblers. Additional methods: https://github.com/david-toews/vermivora_amplicon