ARCHIVED: Maximum likelihood analysis of phylogenetic data

Note: Phylogeny is defined as the evolutionary tree or lines of descent of living species.

Maximum likelihood methods of statistical inference were first developed in the 1930s by R.A. Fisher. Theoretical application to phylogenetic analysis was developed by Joseph Felsenstein in the 1970s and early 1980s. Maximum likelihood methods of phylogenetic inference are superior to some other methods, particularly when the data set includes highly divergent sequences. Such sequences are desirable, but increase the computational difficulty enormously. Parallel computing methods now make the analysis of such large data sets practical.

The program fastDNAml (Olsen et.al. 1994, based on Felsenstein 1981) computes the likelihood of various phylogenetic trees, starting with aligned DNA sequences from a number of species.

Indiana University has modified and extended the serial version of fastDNAml to run in parallel on heterogenous and widely distributed systems. More information on that demonstration is available.

The activities of this group support and have been funded in part by the National Science Foundation (NSF) under Grant No. 0116050 and Grant CDA-9601632; Shared University Research Grants from International Business Machines, Inc. (IBM); and the Indiana Genomics Initiative (INGEN). The Indiana Genomics Initiative of Indiana University is supported in part by Lilly Endowment. Inc. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of NSF, IBM, or Lilly Endowment Inc.