Functional characterization of miRNAs in prostate cancer using functional protein networks

The cell is a highly organized system of interacting molecules including DNA, proteins, mRNAs and miRNAs. Analyzing the cell from a systems perspective by integrating different types of data helps reveal the complexity behind the biological process. Although there is emerging evidence that some microRNAs can function as oncogenes or tumour suppressors, the role of microRNAs in mediating cancer progression and metastasis remains not fully explored. In this work we integrate functional protein networks to characterize the downstream effect of miRNA-mRNA interaction. The proposed method infers a novel miRNA-mRNA regulatory network, not only based on the correlation between miRNA and their targets, but also by taking into account the down stream effects of miRNA-mRNA interaction on interacting proteins. In here we build a model that integrates mRNA and miRNA expression data, a predicted miRNA-mRNA regulatory network (based on sequence complementary) and functional protein interaction data to better characterize the functional role of individual miRNAs. The network is used to predict the effect of miRNA on cancer progression. The model is then applied to two prostate cancer datasets that include normal, primary and metastatic cancer samples to identify prognostic miRNAs that are associated with cancer progression. Network based model demonstrated promising to predict cancer recurrence and death. Furthermore, the model is used to predict miRNA that have similar mode of action to drugs. Results revealed that the proposed method identify miRNA-mRNA modules that play a key role in prostate cancer progression and they are promising targets as prognostic biomarkers.