Phylogenetic and phylogeographic studies rely on the accurate quantification of biodiversity. In recent studies of taxonomically ambiguous groups, species boundaries are often determined based on multi-locus sequence ...Phylogenetic and phylogeographic studies rely on the accurate quantification of biodiversity. In recent studies of taxonomically ambiguous groups, species boundaries are often determined based on multi-locus sequence data. Bayesian Phylogenetics and Phylogeography （BPP） is a coalescent-based method frequently used to delimit species; however, empirical studies suggest that the requirement of a user-specified guide tree biases the range of possible outcomes. We evaluate fifteen multi-locus datasets using the most recent iteration of BPP, which eliminates the need for a user-specified guide tree and reconstructs the species tree in synchrony with species delimitation （= unguided species delimitation）. We found that the number of species recovered with guided versus unguided species delimitation was the same except for two cases, and that posterior probabilities were generally lower for the unguided analyses as a result of searching across species trees in addition to species delimitation models. The guide trees used in previous studies were often discordant with the species tree topologies estimated by BPP. We also compared species trees estimated using BPP and ＊BEAST and found that when the topologies are the same, BPP tends to give higher posterior probabilities [Current Zoology 61 （5）： 866-873, 2015].展开更多
基金We thank the authors of species delimitation studies listed in Table 1 for permission to use their datasets in our study. We also thank J. Grummer for providing feedback on an earlier draft of the manuscript. This work was facilitated though the use of advanced computational, storage, and networking infrastructure provided by the Hyak supercomputer system at the University of Washington. I.W.C.S. was supported by a graduate scholarship from the Consejo Nacional de Ciencia y Tecnologia （CONACYT）, and N.M.B was supported by the National Science Foundation Graduate Research Fellowship （NSF-GRFP）.
文摘Phylogenetic and phylogeographic studies rely on the accurate quantification of biodiversity. In recent studies of taxonomically ambiguous groups, species boundaries are often determined based on multi-locus sequence data. Bayesian Phylogenetics and Phylogeography （BPP） is a coalescent-based method frequently used to delimit species; however, empirical studies suggest that the requirement of a user-specified guide tree biases the range of possible outcomes. We evaluate fifteen multi-locus datasets using the most recent iteration of BPP, which eliminates the need for a user-specified guide tree and reconstructs the species tree in synchrony with species delimitation （= unguided species delimitation）. We found that the number of species recovered with guided versus unguided species delimitation was the same except for two cases, and that posterior probabilities were generally lower for the unguided analyses as a result of searching across species trees in addition to species delimitation models. The guide trees used in previous studies were often discordant with the species tree topologies estimated by BPP. We also compared species trees estimated using BPP and ＊BEAST and found that when the topologies are the same, BPP tends to give higher posterior probabilities [Current Zoology 61 （5）： 866-873, 2015].
