Learning gene network using Bayesian network framework

Abstract

Learning gene networks is one of the central problems in molecular biology. In recent years, with enormous microarray data becoming available, learning gene network has received increasing attention, becoming one of the hottest topics in computational biology. However, the data problem and the complexity of gene regulatory systems make learning difficult. Moreover, some important biological factors are not considered in most published works. There factors include: various time delays among gene regulatory systems, the effects of complexes and the effect of proteins as hidden variables in microarray data. In this thesis, several learning methods based on Bayesian network framework are proposed to take into account these important biological factors: 1) The Time Delayed Network Learning (TDNL) algorithm; 2) the Conditional Dependence (CD) learning algorithm. 3) the Semi-fixed Structure Expectation Maximization (SSEM) algorithm. The effectness of the proposed methods is verified by experiments on both artificial and real-life gene expression data.