Microarray Time Series Modeling and Variational Bayesian Method for Reverse Engineering Gene Regulatory Networks

Abstract

Gene expression is a complex process controlled by underling biological interactions. One model that tries to explain these relationships at a genetic level is the gene regulatory networks. Uncovering regulatory networks are extremely important for live sciences to understand how genes compete and are associated. Despite measurement methods have been successfully developed within the microarray technique, the analysis of genomic data is difficult due to the vast amount of information considered. We address here the problem of modeling the gene regulatory networks by a novel linear model and we propose a Bayesian approach to learn this structure from microarray time series.