|
Activity Number:
|
137
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #306769 |
|
Title:
|
Identifying Activated Molecular Pathways Using Bayesian Methods
|
|
Author(s):
|
Yifang Zhao*+ and Lynn Kuo and Dong-Guk Shin and Fang Yu
|
|
Companies:
|
University of Connecticut and University of Connecticut and University of Connecticut and University of Connecticut
|
|
Address:
|
215 Glenbrook Road, U-4120, Storrs, CT, 06269-4120,
|
|
Keywords:
|
pathway ; Bayes factor ; MCMC
|
|
Abstract:
|
Identifying molecular pathways most activated in a defined stage of cell differentiation or when cells are exposed to environmental stimuli provides more insights on the functional information about genes. We propose novel methods to evaluate a set of possible pathways obtained from KEGG or BioCarta databases on their activations from the microarray and proteomic data. These high-throughput data are further supplemented by prior information constructed from literature search on the gene-to-gene promotion or inhibition knowledge. The Bayes factor approach is used to evaluate the evidence for each activated pathway. Essentially, we develop Markov chain Monte Carlo methods and Bayesian model selection methods to identify a set of pathways most activated by observing the high-throughput data.
|
