Year: 2009
Type: article
Abstract: Central to the study of Life is knowledge both about the underlying relationships \namong living things and the processes that have molded them into their diverse forms. \nPhylogenetics provides a powerful toolkit for investigating both aspects. Bayesian \nphylogenetics has gained much popularity, due to its readily interpretable notion of \nprobability. However, the posterior probability of a phylogeny, as well as any dependent \nbiological inferences, is conditioned on the assumed model of evolution and its priors, \nnecessitating care in model formulation. In Chapter 1, I outline the Bayesian perspective \nof phylogenetic inference and provide my view on its most outstanding questions. I then \npresent results from three studies that aim to (i) improve the accuracy of Bayesian \nphylogenetic inference and (ii) assess when the model assumed in a Bayesian analysis is \ninsufficient to produce an accurate phylogenetic estimate. As phylogenetic data sets increase in size, they must also accommodate a greater \ndiversity of underlying evolutionary processes. Partitioned models represent one way of \naccounting for this heterogeneity. In Chapter 2, I describe a simulation study to \ninvestigate whether support for partitioning of empirical data sets represents a real signal \nof heterogeneity or whether it is merely a statistical artifact. The results suggest that \nempirical data are extremely heterogeneous. The incorporation of heterogeneity into \ninferential models is important for accurate phylogenetic inference. \nBayesian phylogenetic estimates of branch lengths are often wildly unreasonable. \nHowever, branch lengths are important input for many other analyses. In Chapter 3, I \nstudy the occurrence of this phenomenon, identify the data sets most likely to be affected, \ndemonstrate the causes of the bias, and suggest several solutions to avoid inaccurate \ninferences. \nPhylogeneticists rarely assess absolute fit between an assumed model of evolution \nand the data being analyzed. While an approach to assessing fit in a Bayesian framework \nhas been proposed, it sometimes performs quite poorly in predicting a model’s \nphylogenetic utility. In Chapter 4, I propose and evaluate new test statistics for assessing \nphylogenetic model adequacy, which directly evaluate a model’s phylogenetic \nperformance.
Author: Jeremy M. Brown
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
Funder: National Science Foundation
OpenAlex ID: W2123595455