|
Activity Number:
|
200
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305140 |
|
Title:
|
Bayesian Analysis of Array CGH Data
|
|
Author(s):
|
Xiaowei Wu*+ and Hongxiao Zhu and Marek Kimmel
|
|
Companies:
|
Rice University and The University of Texas M.D. Anderson Cancer Center and Rice University
|
|
Address:
|
6100 Main St. MS-138, Houston, TX, 77005,
|
|
Keywords:
|
Copy number alterations ; Intensity ratios ; Reversible jump
|
|
Abstract:
|
We propose a Bayesian approach to analyze array comparative genomic hybridization (CGH) data. Different from most currently available methods, this new approach builds a Bayesian hierarchical model for the data and use a reversible jump Markov chain Monte Carlo (MCMC) algorithm for posterior sampling. The estimated parameters are used to identify copy number changes (gains/losses) in the target DNA sequence. This method is computationally efficient and is flexible to model various covariance structures of the data. Moreover, it also has great advantage in analyzing recurrent copy number alterations (CNAs) in multiple arrays. Simulation study shows a good performance of the proposed method. As a real data analysis example, we apply the method to publicly available Corriel cell lines data and obtain satisfying results.
|
