Research Article

Identification of Prognostic Risk Model Based on DNA Methylation-Driven Genes in Esophageal Adenocarcinoma

Figure 5

Prognostic risk model based on EAC MDGs. (a) The choice of the adjustment parameter λ by Lasso-penalized Cox regression analysis. (b) Each curve represents a MDG. Tenfold cross-validation was utilized to calculate optimal lambda, which leads to minimum mean cross-validation error. The vertical axis represents the mean-square error, while the horizontal axis represents the value of log (λ). (c) EAC patients were divided into two groups according to the risk model score, and Kaplan-Meier survival curves were compared by log-rank test. . (d) ROC curve of the 3-year overall survival rate of the risk model.
(a)
(b)
(c)
(d)