|
Activity Number:
|
525
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #303573 |
|
Title:
|
Using Dynamic Bayesian Networks with Hidden States to Infer Gene Regulatory Networks
|
|
Author(s):
|
Andrea Rau*+ and Florence Jaffrézic and Jean-Louis Foulley and Rebecca W. Doerge
|
|
Companies:
|
Purdue University and French National Institute for Agricultural Research and French National Institute for Agricultural Research and Purdue University
|
|
Address:
|
INRA, Department of Statistics, West Lafayette, IN, 47906,
|
|
Keywords:
|
Dynamic Bayesian Networks ; Gene regulatory networks ; Microarrays ; Time series ; Empirical Bayes ; Kalman Filter
|
|
Abstract:
|
Gene regulatory networks loosely refer to the interactions that occur among genes and other cellular products. By measuring changes in gene expression over time, it is possible to infer the topology of a particular gene network. However, because microarray studies typically yield information on thousands of genes over few biological replicates, these systems are quite difficult to elucidate. An approach that has proven to be effective for inferring networks from such data is the Dynamic Bayesian Network. We use an iterative empirical Bayesian procedure with Kalman filter to estimate the posterior distributions of network parameters. We compare our method to similar existing methods on simulated data and a real microarray time series data set. Our method has the potential to identify gene networks with comparable sensitivity and specificity in a reasonable amount of computational time.
|
