Integration of multiple data sources for identifying functional modules using Bayesian network

Data integration is a crucial step in cancer related bioinformatics studies. Bayesian Networks (BNs) is one of the most commonly used methods for integration of multiple data sources. In this paper, we present a modified BN model that can capture and integrate heterogeneous data to increase its predictive performance. The model allows us to infer aberrant networks with scale-free and small world properties, and to group molecules into functional modules or pathways based on the primary function and biological features. Application of this method to gene and protein biomarkers of hepatocellular carcinoma (HCC) led to identification of modules that significantly contribute to HCC development and progression. The modules include cell cycle dysregulation, increased angiogenesis (e.g., vascular endothelial growth factor, blood vessel morphogenesis), oxidative metabolic alterations, and aberrant activation of signaling pathways involved in cellular proliferation, survival and differentiation (e.g., Wnt pathways). The central findings and conclusions derived from our modified BN model are consistent with those previously reported results.