| Evolutionary Similarity Among Genes When Data Are Sparse (2000) | |||||||||||||||||
Abstract | |||||||||||||||||
| Introduction Genes that encode functionally interacting proteins may show similar evolutionary histories to maintain their functional relationships over time (van Valen, 1973). We consider two genes that share similar evolutionary histories (topologies describing the branching patterns and branch lengths along the topologies) to be evolutionary similiar. We use a Bayesian statistical framework to test for evolutionary similarity (ES) between multiple genes implicated in the same biological pathway. A Bayesian approach overcomes several shortfalls found in likelihood based methods. Likelihood based methods neglect that the topologies are discrete and data are sparse, and so standard asymptotics may not apply (Whelan and Goldman, 1999), and these approaches can not make inference unconditional on topology, as the topologies are non-nested. We begin with a common model used to infer the evolutionary history of genetic sequences from multiple organisms. We develop the appropriat | |||||||||||||||||
Publication details | |||||||||||||||||
| |||||||||||||||||