Strange bayes indeed: uniform topological priors imply non-uniform clade priors

Pickett, Kurt M.; Randle, Christopher P.

doi:10.1016/j.ympev.2004.09.001

Cited by 106 publications

(75 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…B in the online edition of the International Journal of Plant Sciences) were constrained to be monophyletic. Bayesian clade support was assessed via posterior probabilities, although we acknowledge current controversy over the reliability of clade support via posterior probabilities (Simmons et al 2004;Pickett and Randle 2005).…”

Section: Phylogeny Reconstruction and Clade Supportmentioning

confidence: 99%

“…This value is more conservative than the 75% used by Chase et al (2000). For Bayesian comparisons, the threshold for support was arbitrarily selected as 0.70, a very conservative measure given that posterior 379 HARDY ET AL.-MORPHOLOGY AND SPECIES-LEVEL PHYLOGENETICS probabilities tend to be higher than the corresponding bootstrap measures (Simmons et al 2004;Pickett and Randle 2005).…”

Section: Congruencementioning

confidence: 99%

See 1 more Smart Citation

A Phylogeny for the African Restionaceae and New Perspectives on Morphology’s Role in Generating Complete Species Phylogenies for Large Clades

Hardy¹,

Moline

Linder

2008

International Journal of Plant Sciences

View full text Add to dashboard Cite

Hardy, C R; Moline, P ; Linder, H P Hardy, C R; Moline, P; Linder, H P (2008 Difficulties with obtaining complete species-level phylogenies include (1) the accurate identification and sampling of species, (2) obtaining a complete species sampling, and (3) resolving relationships among closely related species.We addressed these in a study of 317 species and subspecies of the African Restionaceae. Accurate species identification and collection in the field was facilitated by a morphology-based interactive key to all species. Despite intensive fieldwork, however, material for DNA extraction could not be obtained for 20 of the 292 species of the focal Restio subclade. Furthermore, the 6831 aligned nucleotides and 1685 parsimonyinformative sequence characters were insufficient to resolve relationships fully within the clade. A simulation indicated that an additional 5000-7000 bases may have been needed to achieve supported resolution in the neighborhood of 95%-100%. Instead of further sequencing, we investigated the phylogenetic utility of the large set of characters contained within the interactive key data set, exploiting recent advances in parsimony and Bayesian programs that allow multistate and supermultistate (including continuous for parsimony) morphological characters. On doing so, parsimony resolution increased 17% to nearly 100%, and overall support increased in both parsimony (bootstrap) and Bayesian (posterior probability) frameworks. Taxa for which DNA data were lacking could be placed in fully resolved positions. Experiments using the parsimony ratchet indicated that placement of these morphology-only taxa may have been completely accurate 30% of the time, to within three nodes of accuracy 60% of time, and accurate to genus 96% of the time. Accurate placement of morphology-only taxa through Bayesian analysis may require extensive effort devoted toward exploring tree and parameter space. We conclude that the increasingly available large morphological data sets associated with interactive keys or informatics initiatives represent convenient yet potentially powerful tools in overcoming many of the commonly encountered obstacles in molecular-based species-level phylogenetics. Difficulties with obtaining complete species-level phylogenies include (1) the accurate identification and sampling of species, (2) obtaining a complete species sampling, and (3) resolving relationships among closely related species. We addressed these in a study of 317 species and subspecies of the African Restionaceae. Accurate species identification and collection in the field was facilitated by a morphology-based interactive key to all species. Despite intensive fieldwork, however, material for DNA extraction could not be obtained for 20 of the 292 species of the focal Restio subclade. Furthermore, the 6831 aligned nucleotides and 1685 parsimonyinformative sequence characters were insufficient to resolve relationships fully within the clade. A simulation indicated that an additional 5000-7000 bases may have been needed to achieve s...

show abstract

Section: Phylogeny Reconstruction and Clade Supportmentioning

confidence: 99%

Section: Congruencementioning

confidence: 99%

A Phylogeny for the African Restionaceae and New Perspectives on Morphology’s Role in Generating Complete Species Phylogenies for Large Clades

Hardy¹,

Moline

Linder

2008

International Journal of Plant Sciences

View full text Add to dashboard Cite

show abstract

“…A prior distribution that is uniform over tree topologies is not uniform over labelled histories or clade formations, where the latter inconsistency has been used to (erroneously!) question the validity of the Bayesian approach per se (Pickett and Randle, 2005). As pointed out by Velasco (2008), the ignorance should be expressed in terms of the physical processes that generate the entities of interest.…”

Section: Future Workmentioning

confidence: 99%

Addressing the Shortcomings of Three Recent Bayesian Methods for Detecting Interspecific Recombination in DNA Sequence Alignments

Husmeier¹,

Mantzaris²

2008

Statistical Applications in Genetics and Molecular Biology

View full text Add to dashboard Cite

This version is available at https://strathprints.strath.ac.uk/42512/ Strathprints is designed to allow users to access the research output of the University of Strathclyde. Unless otherwise explicitly stated on the manuscript, Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Please check the manuscript for details of any other licences that may have been applied. You may not engage in further distribution of the material for any profitmaking activities or any commercial gain. You may freely distribute both the url (https://strathprints.strath.ac.uk/) and the content of this paper for research or private study, educational, or not-for-profit purposes without prior permission or charge.Any correspondence concerning this service should be sent to the AbstractWe address a potential shortcoming of three probabilistic models for detecting interspecific recombination in DNA sequence alignments: the multiple changepoint model (MCP) of Suchard et al. (2003), the dual multiple change-point model (DMCP) of Minin et al. (2005), and the phylogenetic factorial hidden Markov model (PFHMM) of Husmeier (2005). These models are based on the Bayesian paradigm, which requires the solution of an integral over the space of branch lengths. To render this integration analytically tractable, all three models make the same assumption that the vectors of branch lengths of the phylogenetic tree are independent among sites. While this approximation reduces the computational complexity considerably, we show that it leads to the systematic prediction of spurious topology changes in the Felsenstein zone, that is, the area in the branch lengths configuration space where maximum parsimony consistently infers the wrong topology due to long-branch attraction. We apply two Bayesian hypothesis tests, based on an inter-and an intra-model approach to estimating the marginal likelihood. We then propose a revised model that addresses these shortcomings, and compare it with the aforementioned models on a set of synthetic DNA sequence alignments systematically generated around the Felsenstein zone.

show abstract

“…Simulation studies have expressed concerns about evolution rate priors 17 and the use of uninformative priors in Bayesian analyses [66][67] and, therefore, the choice of priors in any Bayesian framework should be carefully considered. Furthermore, whether posterior probabilities are a reliable source of clade confidence and if alternative measures would be preferable 68 should be considered whenever estimating species trees under a Bayesian framework.…”

Section: Reviewmentioning

confidence: 99%

Species tree inference in the age of genomics

Whelan

2011

Trends Evol Biol

View full text Add to dashboard Cite

Species trees are an essential tool in conservation and evolutionary biology. In phylogenomics, not only is data choice (e.g. using unlinked orthologs rather than paralogs) an important systematic consideration, but the choice of phylogenetic algorithm is also important. Since individual gene phylogenies can differ from the true species phylogeny, new methods have been proposed for species tree estimation using multiple unlinked genes. Improvements in genome sequencing technologies have increased the amount of data available to researchers and this has increased the utility of multi-locus species tree inference methods. The Bayesian methods BEST and *BEAST that incorporate a coalescent model to account for gene tree and species tree conflict offer promising advances in species tree inference directly from DNA sequences. Methods that infer species trees from gene trees rather than directly from sequence data such as STAR, STEAC, NJ st and the likelihood method STEM have been recently developed as computationally efficient alternatives. Bayesian concordance analysis, which has been shown to perform well when horizontal gene transfer is the cause of gene tree and species tree conflict, is also discussed. Furthermore, methods for species delimitation including a non-parametric species tree inference method that does not require a priori species assignments can remove subjectivity from species delimitation. Here, I review the assumptions, required inputs, and the performance of these methods under simulation and in recent empirical studies. Researchers in many disciplines should understand the methods available for phylogenomic species tree inference in order to enhance evolutionary and conservation studies.

show abstract

Strange bayes indeed: uniform topological priors imply non-uniform clade priors

Cited by 106 publications

References 49 publications

A Phylogeny for the African Restionaceae and New Perspectives on Morphology’s Role in Generating Complete Species Phylogenies for Large Clades

A Phylogeny for the African Restionaceae and New Perspectives on Morphology’s Role in Generating Complete Species Phylogenies for Large Clades

Addressing the Shortcomings of Three Recent Bayesian Methods for Detecting Interspecific Recombination in DNA Sequence Alignments

Species tree inference in the age of genomics

Contact Info

Product

Resources

About