|
Activity Number:
|
511
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #303655 |
|
Title:
|
Using a Dirichlet Process Mixture of Hidden Markov Models for Protein Conformation Angle Data
|
|
Author(s):
|
Kristin P. Lennox*+ and David B. Dahl and Marina Vannucci and Ryan Day and Jerry Tsai
|
|
Companies:
|
Texas A&M University and Texas A&M University and Rice University and University of the Pacific and University of the Pacific
|
|
Address:
|
Department of Statistics, College Station, TX, 77843-3143,
|
|
Keywords:
|
Bayesian nonparametrics ; Density estimation ; Protein structure prediction ; Torsion angles ; Von Mises distribution
|
|
Abstract:
|
Bayesian models for protein torsion angle data have given new insights into the behavior of these angle pairs. One area of interest that has not been previously addressed is the joint modeling of angle pairs at multiple sequence positions. Such modeling is difficult due to the fact that the number of protein structures available to estimate distributions of interest is typically small, and because not all proteins have angle pairs at each sequence position. We propose a new semiparametric model that copes with this ``sparse data" problem by leveraging known information about protein secondary structure. We demonstrate our technique by modeling a loop structure in the globins family of proteins, and compare our results with those given by other loop modeling techniques.
|
