Identification of Transcription Factor/Gene Axis in Colon Cancer Using a Methylome Approach.

Colon cancer is one of the most commonly diagnosed cancers worldwide. Both environmental and molecular characters can influence its development. DNA methylation has been heralded as a promising marker for use in cancer prevention, diagnosis, and treatment. It has been shown to facilitate cancer progression through multiple mechanisms. Changes in DNA methylation can inhibit or promote the binding of transcription factors (TFs) and further disturb gene regulation. Detection of DNA methylation-mediated regulatory events in colon cancer are critical for mining novel biomarkers. Here, we explore the influence of CpG sites located at promoter regions of differentially expressed genes and identify methylation-gene relationships using expression-methylation quantitative trait loci. We find that promoter methylation sites mainly negatively regulate the corresponding genes. We also identify candidate TFs that can bind to these sites in a sequence-dependent manner. By integrating transcriptome and methylome profiles, we construct a TF-CpG-gene regulatory network for colon cancer, which is used to determine the roles of TFs and methylation in the transcription process. Finally, based on TF-CpG-gene relationships, we design a framework to evaluate patient prognosis, which shows that one TF-CpG-gene triplet is significantly associated with patient survival rate and represents a potential novel biomarker for use in colon cancer prognosis and treatment.