Considering gene flow when using coalescent methods to delimit lineages of North American pitvipers of the genusAgkistrodon

Abstract

Examining species diversity and mechanisms of speciation using coalescent models provides a framework for how regional diversity is accrued, even in well-studied areas such as the Nearctic. It is likely, that gene flow among closely-related species with adjacent distributions may be common. However, the absence of gene flow is a primary assumption of many phylogeographical methods that produce species trees and delimit species using Bayesian or likelihood functions in a coalescent framework. In the present study, we examine delimitation when gene flow between species is present using empirical datasets from two species of North American pitvipers of the genus Agkistrodon. We also use niche modelling to determine whether these young lineages occur in distinct environmental niches. To manage the problem of gene flow between species, we first identify admixed individuals, demonstrate that gene flow has occurred, and then identify the impact of alternative population assignments of admixed individuals on delimitation posterior probabilities. In addition, we examine the influence of mitochondrial genes relative to other loci combined in coalescent analyses that delimit species. Here, we find that the copperheads (Agkistrodon contortrix) and the cottonmouths (Agkistrodon piscivorus) are each composed of two distinct species, with each occupying different niches. Importantly, we find that species can be delimited when the amount of gene flow between lineages is low, although the methods are acutely sensitive to population assignment of individuals.

Biogeographic barriers, Pleistocene refugia, and climatic gradients in the southeastern Nearctic drive diversification in cornsnakes (Pantherophis guttatus complex)

“…The turnover in habitats and environments across the southeastern Nearctic is an important determinant of population genetic structure (Burbrink & Guiher, ; McKelvy & Burbrink, ) and influences the assemblage of taxa into communities (Burbrink & Myers, ). Here we find significant ecological differentiation between both sister lineage pairs, which likely lead to ecologically‐mediated population differentiation and speciation within this complex.…”

Species delimitation with gene flow: A methodological comparison and population genomics approach to elucidate cryptic species boundaries in Malaysian Torrent Frogs

“…Model‐based methods make simplifying assumptions about certain parameters (e.g., gene flow, population size) during the speciation process, and range in complexity from noncoalescent, sequence‐based methods that model speciation in terms of number of substitutions (Zhang, Kapli, Pavlidis, & Stamatakis, ) to highly parameterized Bayesian models based on the multispecies coalescent (Yang & Rannala, ) that allow for the integration of multiple data types into a single model‐based framework (Solís‐Lemus, Knowles, & Ané, ). The efficacy of each method depends on how well the model fits the data, and processes that violate model assumptions such as gene flow (Burbrink & Guiher, ; Sousa & Hey, ; Streicher et al., ) or spatial autocorrelation (Meirmans, ; Reeves & Richards, ) can yield inaccurate species delimitations.…”

Divergence of tropical pitvipers promoted by independent colonization events of dry montane Andean habitats

“…Burbrink and Guiher () recently used multilocus data and alternative population assignments of admixed individuals to demonstrate the impact of gene flow on the identification and delimitation of lineages of North American Agkistrodon pitvipers. The approach implemented in our study goes one step further by using treemix and model‐based analyses in fastsimcoal for our population genomic data in order to disentangle the role of gene flow during the origin of montane lineages of terciopelo pitvipers in Ecuador.…”

“…In addition to potentially being affected by incomplete lineage sorting, coalescent-based inferences in systems characterized by rapid speciation events could also be affected by gene flow because methods for lineage identification based on coalescent methods assume that it has ceased upon species formation (Burbrink & Guiher, 2015). Thus, gene flow could reduce the accuracy of species tree inferences, resulting in underestimates of lineage divergence times and/ or inference of an erroneous topology (Gruenstaeudl, Reid, Wheeler, & Carstens, 2015).…”

“…A closer relationship between the two highland lineages can therefore be accounted for by the fact that incompletely sorted ancestral polymorphisms were not taken into account with single-locus analyses or the evaluation of a concatenated multilocus dataset. Burbrink and Guiher (2015)…”