Gene co-expression analysis identifies common modules related to prognosis and drug resistance in cancer cell lines.

Abstract

To discover a common gene co-expression network in cancer cell, we applied weighted gene co-expression network analysis to transcriptional profiles of 917 cancer cell lines. Fourteen biologically meaningful modules were identified, including cytoskeleton, cell cycle, RNA splicing, signaling pathway, transcription, translation and others. These modules were robust in an independent human cancer microarray dataset. Furthermore, we collected 11 independent cancer microarray datasets, and correlated these modules with clinical outcome. Most of these modules could predict patient survival in one or more cancer types. Some modules were predictive of relapse, metastasis and drug resistance. Novel regulatory mechanisms were also implicated. In summary, our findings, for the first time, provide a modular map for cancer cell lines, new targets for therapy and modules for regulatory mechanism of cancer development and drug resistance.