An Information Gain Approach to Infer Gene Regulatory Networks

The inference of a Gene Regulatory Network (GRN) using gene expression data is a major research topic in bioinformatics. Modeling GRNs is significantly important in order to understand gene dependencies, regulatory functions among genes, biological processes, way of process occurrence and avoiding some unplanned processes (disease). Due to the huge number of genes and the small number of samples, reliable inference of GRNs is still a vital challenge and providing efficient inference algorithms is a serious demand. In this paper, a rigorous framework for addressing GRNs inference is introduced. We propose a novel method for GRNs inference using feature selection approach based on information theory (information gain). In addition, by imposing a constraint on the information gain scores, the numbers of false inferred edges have been reduced, dramatically. The experimental results using biological data reveal that in spite of small number of samples and large number of genes, this method has found the gene interactions efficiently. Furthermore, the outcomes demonstrate that the proposed method achieves a comparable accuracy rate to the some state-of-the-art algorithms. Moreover, the sensitivity rate of the proposed method with respect to the other methods is increased 35% (in average).