Identifying key miRNA–mRNA regulatory modules in cancer using sparse multivariate factor regression

Abstract

The interactions between microRNAs (miRNAs) and messenger RNAs (mRNAs) are known to have a major effect on the formation and progression of cancer. In this study, we identified regulatory modules of 32 cancer types using a sparse multivariate factor regression model on matched miRNA and mRNA expression profiles of more than 9,000 primary tumors. We used an algorithm that decomposes the coefficient matrix into two low-rank matrices with separate sparsity-inducing penalty terms on each. The first matrix linearly transforms the predictors to a set of latent factors, and the second one regresses the responses using these factors. Our solution significantly outperformed another decomposition-based approach in terms of normalized root mean squared error in all 32 cohorts. We demonstrated the biological relevance of our results by performing survival and gene set enrichment analyses. The validation of overall results indicated that our solution is highly efficient for identifying key miRNA–mRNA regulatory modules.