| |
Mathematical Genomics |
| April 13, 2009 to April 15, 2009 |
| Organized By: David Galas (Institute for Systems Biology), Richard Olshen (Co-chair) (Stanford University), Rick Woychik (The Jackson Laboratory), Nancy Zhang (Co-chair) (Stanford University) |
| |
| Return to Workshop Description |
| |
A Bayesian Probabilistic Approach Toward Transforming Public Microarray Repositories Into Disease Diagnosis Databases
Tuesday April 14, 2009
12:00PM - 12:45PM
Speakers:
Haiyan Huang
Abstract:
The rapid accumulation of microarray gene expression data has offered unprecedented opportunities to study human diseases. The NCBI Gene Expression Omnibus is currently the largest database that systematically documents the genome-wide molecular basis of diseases. However, thus far, this resource has been far from being fully utilized. In this paper, we report the first study to transform public microarray databases into automated disease diagnosis databases. We developed a systematic framework to interrogate cross-platform microarray data and heterogeneous disease annotations; and more importantly, we designed a two-stage Bayesian probabilistic method based on careful modeling of the complex properties of noisy data with different levels of credence. A high level of overall diagnostic accuracy was demonstrated by cross validation. Moreover, we demonstrate that the power of our method increase significantly with the continued growth of public microarray repositories. Our framework thus provides an important application for the enormous and growing quantity of costly-to-generate, yet freely available, microarray data.
This is a joint work with Dr. X. Jasmine Zhou and Dr. Jim Chun-chi Liu from the University of Sourthern California.
|
