Dissertation Research: The Detection of Recombination from DNA Sequences and its Impact in Phylogeny Reconstruction

点击翻译按钮获取中文摘要

0073154Crandall and Posada A fundamental assumption in molecular-phylogenetic reconstruction is that there is only one history underlying the set of DNA sequences under study. However, different regions of the gene under study can have different evolutionary histories due to reticulate evolution caused by recombination between chromosomes, gene conversion or horizontal transfer. Traditional methods of phylogenetic reconstruction simply ignore recombination, and the impact of recombination on the different phylogenetic estimation algorithms is unknown. Graduate student David Posada, under the direction of Dr. Keith Crandall, proposes to characterize, using computer simulations, the performance of different phylogenetic methods when recombination has occurred.A related problem is the establishment of the presence of recombination in real data sets. Numerous methods have been developed to test for the occurrence of reticulate evolution from a set of aligned DNA sequences, and to determine the bounds of these events. Unfortunately, very little is known about the relative strengths and weaknesses of these methods. Computer simulation will be used to characterize the performance of different methods for detecting recombination. The effects of different rates of recombination, of the size of the recombination unit and position of the recombinational event, of mutation rate, of population size, and of sequence length, will all be studied. The different detection algorithms will also be applied to empirical data sets where recombination has been identified and described, and to mitochondrial data sets, where recombination is supposedly absent. This characterization will help indicate the appropriate tools for detecting recombination and for understanding phylogenies of potentially recombining sequences. Given the increasing number of nuclear genes being sequenced, recombination can no longer be ignored in phylogenetic reconstruction methods.