Integrative analysis with expanded DNA methylation data reveals common key regulators and pathways in cancers

The workflow of the integrative analysis and the EAGLING model. a The multi-omics data of 13 cancers from TCGA were used to identify the genes that are differentially expressed and differentially methylated and also contain somatic mutations (i.e. triple-evidenced genes) in each cancer. Pan-cancer analysis revealed that the triple-evidenced genes shared by a majority of the 13 cancers include many previously identified biomarkers or therapeutic targets. b In the model construction, two features are used to build the logistic regression model: the methylation level of the closest CpG based on 450？K array and the methylation value from the WGBS of the corresponding CpG from the tissue that has the most similar local methylation profile with the site to be predicte