Figure 5: Implication of metabolic regulation of methylation in patient survival.
From: Integrative modelling of tumour DNA methylation quantifies the contribution of metabolism

(a–d) Kaplan–Meier curves are depicted comparing groups of patients wherein cancer gene methylation was predictable (red) or not predictable (black) by the met cycle variables (see Methods). Overall survival in days is plotted in each case and censored subjects are shown by vertical tick marks (Methods). Log-rank test P value between the two groups is reported. Survival analysis results and log-rank test P values are shown for brain, liver, kidney and colon cancers, respectively. (e–h) Survival analysis results as described above are reported for bladder, breast, lung and prostate cancers, respectively. Log-rank test P values showed no significant difference between the ‘predictable’ and ‘not predictable’ groups. NS, not significant. Sample sizes: breast=770; lung=450; liver=374; brain=534; bladder=408; kidney=316; prostate=424; and colon=198.