|
Activity Number:
|
409
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #305598 |
|
Title:
|
Reconstructing Evolutionary Trees Using Amino Acid Substitution Models that Allow Rate Variation to Depend on Spatial Location
|
|
Author(s):
|
Xueliang Pan*+ and Dennis K. Pearl and Liang Liu and Dennis J. Pollack
|
|
Companies:
|
The Ohio State University and The Ohio State University and The Ohio State University and The Ohio State University
|
|
Address:
|
3536 Chowning Court, Columbus, OH, 43220,
|
|
Keywords:
|
phylogeny ; Bayesian model ; site variation ; crystallographic structure
|
|
Abstract:
|
Bayesian inference of phylogeny using MCMC methods has been used increasingly over the last decade. As many researchers recognize that sites do not evolve independently at the same rate, more realistic models that allow site variation are now the norm. The most used is the discrete gamma model proposed by Yang. In this study, a model allowing site variation based on its spatial location is proposed. The amino acid sites were partitioned according to their spatial location, obtained from the crystallographic structure; the rate variation of each partition was estimated by Mrbayes. The convergence and quality of this model were evaluated and found to be superior to the gamma model in terms of computational time and likelihood. This study is based on a dataset regarding phosphoglycerate kinase (PGK), an enzyme important in fermentation that is found in species from all kingdoms.
|
