Ultraconserved elements (UCEs) illuminate the population genomics of a recent, high-latitude avian speciation event

• Published in: PeerJ
• Main Authors: Winker, Kevin, Glenn, Travis C., Faircloth, Brant C.
• Other Authors: National Science Foundation, National Center for Genome Analysis Support (Trinity on Galaxy for assembly) and the Research Computing Systems of the University of Alaska Fairbanks (for analyses) provided computational support
• Format: Article in Journal/Newspaper
• Language: English
• Published: PeerJ 2018
• Subjects: General Agricultural and Biological Sciences; General Biochemistry, Genetics and Molecular Biology; General Medicine; General Neuroscience; Plectrophenax nivalis
• Online Access: http://dx.doi.org/10.7717/peerj.5735; https://peerj.com/articles/5735.pdf; https://peerj.com/articles/5735.xml; https://peerj.com/articles/5735.html

Using a large, consistent set of loci shared by descent (orthologous) to study relationships among taxa would revolutionize among-lineage comparisons of divergence and speciation processes. Ultraconserved elements (UCEs), highly conserved regions of the genome, offer such genomic markers. The utility of UCEs for deep phylogenetics is clearly established and there are mature analytical frameworks available, but fewer studies apply UCEs to recent evolutionary events, creating a need for additional example datasets and analytical approaches. We used UCEs to study population genomics in snow and McKay’s buntings ( Plectrophenax nivalis and P. hyperboreus ). Prior work suggested divergence of these sister species during the last glacial maximum (∼18–74 Kya). With a sequencing depth of ∼30× from four individuals of each species, we used a series of analysis tools to genotype both alleles, obtaining a complete dataset of 2,635 variable loci (∼3.6 single nucleotide polymorphisms/locus) and 796 invariable loci. We found no fixed allelic differences between the lineages, and few loci had large allele frequency differences. Nevertheless, individuals were 100% diagnosable to species, and the two taxa were different genetically ( F ST = 0.034; P = 0.03). The demographic model best fitting the data was one of divergence with gene flow. Estimates of demographic parameters differed from published mtDNA research, with UCE data suggesting lower effective population sizes (∼92,500–240,500 individuals), a deeper divergence time (∼241,000 years), and lower gene flow (2.8–5.2 individuals per generation). Our methods provide a framework for future population studies using UCEs, and our results provide additional evidence that UCEs are useful for answering questions at shallow evolutionary depths.