Isolation-driven divergence: speciation in a widespread North American songbird (Aves: Certhiidae)

Abstract

Lineage, or true “species,” trees may differ from gene trees because of stochastic processes in molecular evolution leading to gene-tree heterogeneity. Problems with inferring species trees due to excessive incomplete lineage sorting may be exacerbated in lineages with rapid diversification or recent divergences necessitating the use of multiple loci and individuals. Many recent multilocus studies that investigate divergence times identify lineage splitting to be more recent than single locus studies, forcing the revision of biogeographic scenarios driving divergence. Here we use 21 nuclear loci from regional populations to reevaluate hypotheses identified in an mtDNA phylogeographic study of the Brown Creeper (Certhia americana), as well as identify processes driving divergence. Nuclear phylogeographic analyses identified hierarchical genetic structure, supporting a basal split at roughly 32°N latitude, splitting northern and southern populations, with mixed patterns of genealogical concordance and discordance between datasets within the major lineages. Coalescent-based analyses identify isolation, with little to no gene flow, as the primary driver of divergence between lineages. Recent isolation appears to have caused genetic bottlenecks in populations in the Sierra Madre Oriental and coastal mountain ranges of California, which may be targets for conservation concerns.

Chromosomal patterns of diversity and differentiation in creepers: a next-gen phylogeographic investigation of Certhia americana

“…Because it is a songbird, sequence data may be BLASTed to the Zebra Finch genome to identify upon which chromosome loci are found. There are two major lineages, split at 32°N latitude, within C. americana identified with both mitochondrial (mtDNA; Manthey et al, 2011a) and nuclear DNA (nDNA; Manthey et al, 2011b). Between the major lineages, there is also apparent quicker differentiation and reduced gene flow on the Z chromosome relative to autosomal loci (Manthey and Spellman, 2014).…”

“…Tissue samples of 41 C. americana individuals were obtained from eight localities (Figure 1, Table 1, Supplementary Table 1), representing the structured clades recovered in previous mtDNA analyses (Manthey et al, 2011a). Total genomic DNA was extracted from tissue samples using a QIAGEN (Hilden, Germany) DNeasy tissue extraction kit following the manufacturer protocols.…”

“…With new methods for obtaining reduced-representation libraries of the genome (for example, restriction digest-based methods, Miller et al, 2007 andultraconserved elements, Faircloth et al, 2012) for many individuals, phylogeographic studies may contain thousands of loci with dozens of individuals sampled. In songbirds, strong patterns of interchromosomal synteny (Kawakami et al, 2014), the published genome of the Zebra Finch (Taeniopygia guttata; Warren et al, 2010) and thousands of genetic markers across the genome allow the opportunity to investigate not only phylogeographic structure in a clade well known for its high levels of geographic differentiation (Manthey et al, 2011a), but also diversity and differentiation across chromosomes (Manthey and Spellman, 2014).…”

Conservation genomics reveals multiple evolutionary units within Bell’s Vireo (Vireo bellii)

“…Furthermore, K = 2, 3, 4 all received qualitatively similar probability scores : -145,513.0, -144,494.9, -144,334.2, respectively. In all population models, Texas individuals (groups D and E; (Farrington et al 2014;Hart et al 2014;Manthey et al 2011) is to perform separate analyses on different geographic regions. Thus, we further analyzed population structure by completing analyses for the eastern and western sub-clades with K = 1-4, and we found K = 2 and K = 3 received the best likelihood scores ( (Table 2).…”

Fluctuating fire regimes and their historical effects on genetic variation in an endangered shrubland specialist

“…We coded our nuclear loci as SNPs (single nucleotide polymorphisms; following Manthey et al. 2011) to calculate K with values between 1 and 20 clusters using the admixture (ancient panmixia) and nonadmixture models (ancient fragmentation) with sampling locality as a prior and both correlated and uncorrelated allele frequencies (similar results therefore only correlated frequencies reported), 20 repetitions per cluster with 1 × 10 6 iterations and a burn‐in of 100,000 steps to ensure reproducibility (Gilbert et al. 2012), and a fixed lambda value (which was inferred by setting K  =   1 and allowing lambda to be estimated in an initial analysis).…”