Computational Inferring of Risk Subpathways Mediated by Dysfunctional Non-coding RNAs

Abstract

Non-coding RNAs mediated core elements of pathways contributes to the disorder of biological function in diseases. Identification of non-coding RNAs mediated subpathways not only can help for deciphering the pathogenic mechanism of complex diseases, but also can gain insight into the functional roles of non-coding RNAs in human diseases. Here, we summarized the general steps for identifying non-coding RNA mediated subpathways and overviewed two of our previously developed methods, Subpathway-GMir and Subpathway-LNCE, which were designed to identify miRNAs and lncRNAs mediated risk subpathways respectively. We identified the key subpathway regions by integrating non-coding RNA-target gene associations, interesting genes and non-coding RNAs and pathway topologies. By applying methods to several disease datasets, we confirmed that our methods is effective in identifying risk subpathways and also can help uncover key non-coding RNAs in diseases. Additionally, reproducibility and robustness analysis demonstrated our methods are reliable.

Keywords: Integration; Network; Non-coding RNA; Pathway topology; Subpathway.