A Novel Clinical Six-Flavoprotein-Gene Signature Predicts Prognosis in Esophageal Squamous Cell Carcinoma

Figure 1

Identification of a flavoprotein signature predictive of overall survival of patients with ESCC. (a) Schematic diagram of the study. (b) Random forest supervised classification algorithm. (c) Procedure for identifying the final signature. The accuracies of all 1023 signatures were calculated, and the nine highest accuracies for k = 1, 2, …, 10 are shown in the plot and ROC for the flavoprotein signature prognostic model in the training dataset. (d) Time-dependent ROC for the flavoprotein signature prognostic model at 3–6 years in the training dataset. (e, f) Kaplan–Meier survival curves of patients classified into high- and low-risk groups, using the flavoprotein signature in the training and independent validation datasets. P values were calculated by the log-rank test.

(a)

(b)

(c)

(d)

(e)

(f)