Bayesian networks have become a commonly used tool for inferring structure of gene regulatory networks from gene expression data. In this framework, genes are mapped to nodes of a graph, and Bayesian techniques are used to determine a set of edges that best explain the data, that is, to infer the underlying structure of the network. This chapter begins with an explanation of the mathematical framework of Bayesian networks in the context of reverse engineering of genetic networks. The second part of this review discusses a number of variations upon the basic methodology, including analysis of discrete vs. continuous data or static vs. dynamic Bayesian networks, different methods of exploring the potentially huge search space of network structures, and the use of priors to improve the prediction performance. This review concludes with a discussion of methods for evaluating the performance of network structure inference algorithms.
Comparative evaluation of reverse engineering gene regulatory networks with relevance networks, graphical gaussian models and bayesian networks
