Skip to main content
Fig. 3 | Clinical Epigenetics

Fig. 3

From: Cross talk between acetylation and methylation regulators reveals histone modifier expression patterns posing prognostic and therapeutic implications on patients with colon cancer

Fig. 3

Consensus clustering of histone modification regulators in patients with colon cancer. A Heatmaps demonstrating the histone modifier expression patterns in patients with colon cancer identified by the unsupervised clustering analysis of 88 histone modification regulators in the integrated meta-GEO and TCGA-COAD cohort. Cohort details and histone modification clusters are used as patient annotations. We specifically labeled the names of histone modification regulators highly expressed in patients in the HMC4 cluster. Red values indicate significantly higher expression in the fluorouracil-nonresponse group than in the fluorouracil-response group, blue values indicate significantly higher expression in the fluorouracil-response group than in the fluorouracil-nonresponse group, and gray values indicate statistically insignificant differences. B and C Kaplan–Meier curves of relapse-free survival (B) and overall survival (C) according to histone modifier expression patterns in the meta-GEO cohort. D Kaplan–Meier curves of overall survival according to histone modifier expression patterns in the TCGA-COAD cohort. E Forest plots of the association between HMC4 pattern and overall survival in subgroups stratified by adjuvant chemotherapy conduction. F Forest plots of benefits of adjuvant chemotherapy in different HMC clusters in the GSE39582 and TCGA-COAD cohorts. Unadjusted hazard ratios (boxes) and 95% confidence intervals (horizontal lines) are depicted; G bar charts summarize the proportions of patients with fluorouracil-response signatures and those with nonresponse signatures within and across different histone modifier expression patterns. RFS, relapse-free survival; OS, overall survival; CI, confidence interval; ADJC, adjuvant chemotherapy; R, response; and NR, nonresponse

Back to article page