Phylogenetic trees are important in biology since their applications range from determining protein function to understanding the evolution of species. Maximum Likelihood (ML) is a popular optimization criterion in phylogenetics. However, inference of phylogenies with ML is NP-hard. Recursive-Iterative-DCM3 (Rec-I-DCMS).is a divideand-conquer framework that divides a dataset into smaller subsets (subproblems), applies an external base method to infer subtrees, merges the subtrees into a comprehensive tree, and then refines the global tree with an external global method. In this study we present a novel parallel implementation of Rec-I-DCM3 for inference of large trees with ML. Parallel-Rec-I-DCM3 uses RAxML as external base and global search method. We evaluate program performance on 6 large real-data alignments containing 500 up to 7.769 sequences. Our experiments show that P-Rec-I-DCM3 reduces inference times and improves final tree quality over sequential Rec-I-DCM3 and stand-alone RAxML. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Du, Z., Stamatakis, A., Lin, F., Roshan, U., & Nakhleh, L. (2005). Parallel divide-and-conquer phylogeny reconstruction by maximum likelihood. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3726 LNCS, pp. 776–785). https://doi.org/10.1007/11557654_88
Mendeley helps you to discover research relevant for your work.