Discovering an Integrated Network in Heterogeneous Data for Predicting lncRNA-miRNA Interactions

Long noncoding RNAs (lncRNAs) belong to a class of non-protein coding RNAs, which have recently been found to potentially act as a regulatory molecule in some important biological processes. MicroRNAs (miRNAs) have been proved by many biomedical studies to be closely associated to many human diseases. Recent studies have suggested that lncRNAs could potentially interact with miRNAs to modulate their regulatory roles. Hence, predicting lncRNA-miRNA interactions is biologically significant due to their potential roles in determining the effectiveness of gene regulations. As diverse heterogeneous datasets for describing lncRNA and miRNA have been made available, it becomes more feasible for us to develop a model to describe potential interactions between lncRNAs and miRNAs. In this work, we presented a new computational pipeline, called INLMI, to predict lncRNA-miRNA interactions by integrating the expression similarity network and the sequence similarity network. Based on a measure of similarities between these networks, INLMI computes an interaction score for a pair of lncRNA and a miRNA. The novelty of INLMI lies in that we used network integration on two similarity networks. Using a real data set, we have shown that INLMI can be a very effective approach as the model that it has learnt can be used to very accurately predict lncRNA-miRNA interactions.