Discovery of Functional miRNA-mRNA Regulatory Modules with Computational Methods

MicroRNAs (miRNAs) are a group of single-stranded, non-coding RNAs. They target protein coding mRNAs through complementary base-pairing for cleavage, repressing translation and causing protein degradation. Increasing number of evidences indicate that miRNAs play important roles in cell differentiation, proliferation, growth, mobility, and apoptosis. Consequently, dysregulation of miRNA may lead to human diseases, including cancer. Identifying miRNAs and their target mRNAs, and further building their regulatory networks may give new insights to biological procedures. Several computational methods have been proposed for miRNA studies. Previous works largely focus on the genome-wide discovery of miRNAs and prediction of putative target mRNAs. However, most target prediction methods search different parts of the miRNA-target space heuristically with different criteria leading to different results. In addition, the computational algorithms used to predicate miRNA targets have limited accuracy. Therefore, empirical methods are in need to dig out true miRNA targets and identify the functional regulatory networks involving miRNAs and their target mRNAs. It is crucial for understanding the regulatory mechanism of miRNAs in complex cellular systems. Answering how miRNAs associate with different conditions by regulating their bona fide target mRNAs may solve several problems in miRNA research potentially.

This study proposes a computational method to discover the functional miRNA-mRNA regulatory modules (FMRMs), that is, groups of miRNAs and their target mRNAs that are believed to participate cooperatively in post-transcriptional gene regulation under specific conditions. This method takes advantage of both miRNA target information and expression profiles of miRNAs and mRNAs.

Computational methods by incorporating heterogeneous information may help identify functional regulatory miRNA-mRNA modules. The FMRMs identified in this study include the negatively correlated miRNA-mRNA pairs which associate with prostate cancer and normal condition. They display many meaningful discoveries supported by GO and the literature. By associating miRNA-mRNA pairs with conditions, it potentially can identify the biologically relevant targets of miRNAs and chains of 'miRNA → target gene → condition'. It will give new insight to the biological procedures at the molecular level.

Contact Person: Bing Liu