Targeted Sampling and Target Capture: Assessing Phylogeographic Concordance with Genome-wide Data

Comparative phylogeography provides the necessary framework to examine the factors influencing population divergence, persistence, and change over time. Avise (2000) outlined four aspects of concordance that result when data exhibit significant phylogeographic signal: concordance among sites within a locus, among multiple loci within a species, among multiple species within a region, and between genetic patterns and established biogeographic provinces. To fully address each aspect of concordance, we combined target capture of a set of orthologous loci with targeted geographic sampling of multiple species, thus removing any variability introduced by using different genetic markers and heterogeneous sampling distributions. We used hybrid enrichment and high-throughput sequencing of four anuran species sampled from 36 congruent localities in the Southeastern United States Coastal Plain, a region that represents one of the classic systems in phylogeography. In total, we recovered >375 of the same nuclear loci across species and assembled mitochondrial genomes, resulting in one of the most comprehensive comparative phylogeographic datasets in any region or taxon to date. We used these data to evaluate concordance, compare genetic structure across species, and test previously described biogeographic features in the region including major river drainages and suture zones. We then applied a recently-developed framework to quantify concordance across species using phylogeographic concordance factors. For the four species examined, which have higher dispersal and potentially limited structure compared to many amphibians, we found poor resolution in individual nuclear gene trees even with long (similar to 1400 bp) nuclear sequences. The mitochondrial and multi-locus nuclear datasets, however, produced similar patterns within species, indicated high discordance among species, and suggested little correspondence of genetic patterns with putative biogeographic barriers. Variation in the phylogeographic structure detected may be related to differences in natural history, in that the two habitat generalists exhibited less structure. Our study demonstrates the utility of combining target capture, which is highly repeatable and produces comparable datasets, with a targeted sampling strategy to quantify phylogeographic concordance across diverse taxa in a region with a complex history.