Skip to main content
. 2021 Nov 11;2021:4251763. doi: 10.1155/2021/4251763

Figure 2.

Figure 2

Establishment of an immune gene-based signature for predicting prognosis of gastric cancer in the training set. (a) Selection of variables in LASSO regression model by tenfold cross-verification. (b) Partial likelihood deviance for each λ in LASSO regression model. (c) Ranking of risk scores for gastric cancer patients. (d) Survival status of each patient. (e) Kaplan-Meier curves for OS between the high- and low-risk groups. (f) ROCs of one-, three-, and five-year survival based on this signature. ROCs of one-, three-, and five-year survival based on the immune-related signature constructed by (g) Li et al. and (h) Tian et al. (i) Univariate and (j) multivariate Cox regression analysis of the risk score and other clinicopathological characteristics.