Rachel Lockridge Mueller
Assistant Professor
Department of Biology
Colorado State University
Fort Collins, CO 80523
rlm@colostate.edu
(970) 491-6717
|
P h y l o g e n e t i c M e t h o d o l o g y
Molecular Evolution and Phylogenetic Performance—The success of molecular phylogenetic studies depends on the appropriate choice of marker. My research has used mitochondrial genome sequences to compare the phylogenetic performance of genes and to explore the processes of molecular evolution underlying performance differences (Mueller, Syst. Biol., 2006). I measured the effects of absolute rate of molecular evolution, among-site rate heterogeneity (a), and gene length on phylogenetic performance using ordinal logistic regression. My results (1) show significant effects of both rate and length, but not a, on performance, and (2) identify problems with cox1 as a universal DNA barcode, for which it is being used. I am currently extending this analysis to include multiple taxonomic groups and divergence depths, and I would like to eventually target effective nuclear markers as well. Finally, I plan to complement this empirical work with simulations designed to separate the phylogenetic effects of evolutionary rate from the effects of within-site rate heterogeneity (heterotachy).
Partitioning Genomes for Bayesian Phylogenetic Analysis—The success of molecular phylogenetic studies also depends on the appropriate choice of nucleotide substitution model. I have explored techniques for analyzing whole-genome datasets (Mueller et al. Proc. Nat. Acad. Sci., 2004). I compared four partitioning strategies designed to improve the fit of the substitution model to the data in light of heterogeneous substitution processes among genes and codon positions. My results demonstrate that partitioning by codon position across protein-coding genes is significantly more effective than partitioning by gene, although the latter is more common in published studies.
|
