Abstract:Background. Accumulating evidence demonstrates that complex diseases may arise from cooperative effects of multiple dysfunctional miRNAs. Thus, identifying abnormal functions cooperatively regulated by multiple miRNAs is useful for understanding the pathogenesis of complex diseases.Methods. In this study, we proposed a multistep method to identify dysfunctional miRNA-mRNA regulatory modules (dMiMRMs) in a specific disease, in which a group of miRNAs cooperatively regulate a group of target genes involved in a specific function. We identified dysfunctional miRNAs, which were differentially expressed and inversely regulated most of their target genes, by integrating paired miRNA and mRNA expression profiles and miRNA target information. Then, we identified cooperative functional units, in each of which a pair of miRNAs cooperatively repressed function-enriched and highly interconnected target genes. Finally, the cooperative functional units were assembled into dMiMRMs.Results. We applied our method to glioblastoma (GBM) and identified GBM-associated dMiMRMs at the population, subtype, and individual levels. We identified 5 common dMiMRMs using all GBM samples, 3 of which were associated with the prognosis in patients with GBM and were better predictors of prognosis than were miRNAs or mRNAs alone. By applying our approach to GBM subtypes, we found consistent dMiMRMs across GBM subtypes, and some subtype-specific dMiMRMs were observed. Furthermore, personalized dMiMRMs were identified, suggesting significant individual differences in different patients with GBM.Conclusions. Our method provides the capability to identify miRNA-mediated dysfunctional mechanisms underlying complex diseases.

Identifying Dysfunctional Mirna-Mrna Regulatory Modules by Inverse Activation, Cofunction, and High Interconnection of Target Genes: A Case Study of Glioblastoma