Overcredibility of molecular phylogenies obtained by Bayesian phylogenetics.

Institute of Molecular Evolutionary Genetics and Department of Biology, Pennsylvania State University, 328 Mueller Laboratory, University Park, PA 16802, USA.
Proceedings of the National Academy of Sciences (Impact Factor: 9.81). 01/2003; 99(25):16138-43. DOI: 10.1073/pnas.212646199
Source: PubMed

ABSTRACT Bayesian phylogenetics has recently been proposed as a powerful method for inferring molecular phylogenies, and it has been reported that the mammalian and some plant phylogenies were resolved by using this method. The statistical confidence of interior branches as judged by posterior probabilities in Bayesian analysis is generally higher than that as judged by bootstrap probabilities in maximum likelihood analysis, and this difference has been interpreted as an indication that bootstrap support may be too conservative. However, it is possible that the posterior probabilities are too high or too liberal instead. Here, we show by computer simulation that posterior probabilities in Bayesian analysis can be excessively liberal when concatenated gene sequences are used, whereas bootstrap probabilities in neighbor-joining and maximum likelihood analyses are generally slightly conservative. These results indicate that bootstrap probabilities are more suitable for assessing the reliability of phylogenetic trees than posterior probabilities and that the mammalian and plant phylogenies may not have been fully resolved.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Compared to the striking diversification and levels of endemism observed in many terrestrial groups within the Hawaiian Archipelago, marine invertebrates exhibit remarkably lower rates of endemism and diversification. Supralittoral invertebrates restricted to specific coastal patchy habitats, however, have the potential for high levels of allopatric diversification. This is the case of Ligia isopods endemic to the Hawaiian Archipelago, which most likely arose from a rocky supralittoral ancestor that colonized the archipelago via rafting, and diversified into rocky supralittoral and inland lineages. A previous study on populations of this isopod from O'ahu and Kaua'i revealed high levels of allopatric differentiation, and suggested inter-island historical dispersal events have been rare. To gain a better understanding on the diversity and evolution of this group, we expanded prior phylogeographic work by incorporating populations from unsampled main Hawaiian Islands (Maui, Moloka'i, Lana'i, and Hawai'i), increasing the number of gene markers (four mitochondrial and two nuclear genes), and conducting Maximum likelihood and Bayesian phylogenetic analyses. Our study revealed new lineages and expanded the distribution range of several lineages. The phylogeographic patterns of Ligia in the study area are complex, with Hawai'i, O'ahu, and the Maui-Nui islands sharing major lineages, implying multiple inter-island historical dispersal events. In contrast, the oldest and most geographically distant of the major islands (Kaua'i) shares no lineages with the other islands. Our results did not support the monophyly of all the supralittoral lineages (currently grouped into L. hawaiensis), or the monophyly of the terrestrial lineages (currently grouped into L. perkinsi), implying more than one evolutionary transition between coastal and inland forms. Geometric-morphometric analyses of three supralittoral clades revealed significant body shape differences among them. A taxonomic revision of Hawaiian Ligia is warranted. Our results are relevant for the protection of biodiversity found in an environment subject to high pressure from disturbances.
    PLoS ONE 12/2013; 8(12):e85199. · 3.73 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The greater power of parametric methods over parsimony is frequently observed in empirical phylogenetic analyses by providing greater resolution and higher branch support. This greater power is provided by several different factors, including some that are generally regarded as disadvantageous. In this study we used both empirical and (modified) simulated matrices to examine how Bayesian MCMC, maximum likelihood, and parsimony methods interpret ambiguous optimization of character states. We describe the information content in "redundant" terminals as well as a novel approach to help identify clades that cannot be unequivocally supported by synapomorphies in empirical matrices. Four of our main conclusions are as follows. First, the SH-like approximate likelihood ratio test is a more reliable indicator than the bootstrap of branches that are only ambiguously supported in likelihood analyses wherein only a single fully resolved optimal tree is presented. Second, bootstrap values generated by methods that only ever present a single fully resolved optimal tree are less robust to differences in taxon sampling than are those generated by more conservative methods. Third, PAUP∗ likelihood is more resilient to producing apparently unambiguous resolution and high support from ambiguous characters than is GARLI collapse 1 and MrBayes, which in turn are more resilient than PhyML. GARLI collapse 0, IQ-TREE, and RAxML are the least resilient bootstrapping methods examined. Fourth, frequent discrepancies with respect to resolution and/or branch support may be obtained by methods that only ever present a single fully resolved optimal tree in different contexts that are apparently unique to the specific program and/or method of quantifying branch support.
    Molecular Phylogenetics and Evolution 05/2014; · 4.07 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The supralittoral environment, at the transition between sea and land, is characterized by harsh conditions for life. Nonetheless, evolution of terrestrial isopods (Oniscidea), the only group of Crustacea fully adapted to live on land, appears to have involved a transitional step within the supralittoral. The two most basal oniscidean lineages (Ligiidae and Tylidae) have representatives that successfully colonized the supralittoral. One of them is the genus Tylos, which is found exclusively in supralittoral sandy beaches from tropical and subtropical coasts around the world. Comprehensive phylogenetic hypotheses for this genus are lacking, which are necessary for understanding the evolution and biogeography of a lineage that successfully diversified in the harsh sea-land interface. Herein, we studied the phylogenetic relationships among 17 of the 21 currently recognized species of the genus Tylos, based on sequences from four mitochondrial genes (Cytochrome Oxidase I, Cytochrome b, 16S rDNA, and 12S rDNA). Maximum Likelihood and Bayesian phylogenetic analyses identified several lineages with deep divergences and discrete geographic distributions. Phylogenetic and distributional patterns of Tylos provide important clues on the biogeography and evolution of this group. Large divergences among the most basal clades are consistent with ancient splits. Due to the biological characteristics of Tylos, which likely prevent dispersal of these isopods across vast oceanic scales, we argue that tectonic events rather than trans-oceanic dispersal explain the distribution of Tylos in different continents. Overwater dispersal, however, likely enabled range expansions within some basins, and explains the colonization of volcanic oceanic islands. Present-day distributions were also likely influenced by sea level and climate changes. High levels of allopatric cryptic genetic differentiation are observed in different regions of the world, implying that the dispersal abilities of Tylos isopods are more limited than previously thought. Our results indicate that a taxonomic revision of this group is necessary.
    PLoS ONE 01/2014; 9(4):e94081. · 3.73 Impact Factor

Full-text (2 Sources)

Available from
May 29, 2014