Dynamic Bayesian Network (DBN) with Structure Expectation Maximization (SEM) for Modeling of Gene Network from Time Series Gene

Exploring gene regulatory network is a key topic in molecular biology. In this paper, we present a new dynamic Bayesian network (DBN) framework embedded with structural expectation maximization (SEM) to model gene relationship. It is well-suited for analyzing the time-series data and can deal with cyclical structures that can not be tackled by static Bayesian network. We applied the new method to learning the regulatory network and the metabolic pathway from Saccharomyces Cerevisiae cell cycle gene expression data. The results show that the proposed method is capable of handling missing values in expression data sets, and the inference accuracy can further be improved. Keyword: Microarrays; Gene regulatory networks; Dynamic Bayesian network; Structural expectation maximization

• Type: Conference
• Year: 2006
• Where: BIOCOMP
• Authors: Yu Zhang, Zhidong Deng, Hongshan Jiang, Peifa Jia
• Keywords: Bayesian Network, BIOCOMP 2006, BIOCOMP 2007, Gene Regulatory Networks, Regulatory Networks