Prioritizing Congenital Heart Disease Genes from Transcriptone and Interactome via PageRank

Congenital heart disease (CHD) is a serious disorder with the heart's structure and function presenting at birth. It has become the leading cause of infant morbidity worldwide. Thus, finding the disease genes causing CHD is crucial in clinical diagnosis and medical treatment. Currently, network-based method has become a powerful tool of systematically analyzing complex diseases, including the identification of candidate disease genes. In this paper, we propose a bioinformatics framework of prioritizing CHD genes by leveraging the modified PageRank algorithm on an integrative protein-protein interaction network. Specifically, we weigh the interactions by mutual information of gene expressions for measuring the context specificities between genes. After formulating the network into two components of known disease genes and normal healthy genes, we rank the disease-related genes by bringing the order in the network via PageRank. We conclude that these known disease genes achieve significantly higher ranks compared to the normal genes, and these high-ranked genes perform relevant dysfunctions of CHD risks.