Fig. 3
Construction of a prognostic model in TCGA-HCC. (a) Univariate Cox regression analysis in TCGA-HCC. (b) Coefficient curve. Different colors represent different genes. No zero values were selected as a penalty coefficient (Left). The minimum lambda of the lasso model was selected via 10 folds of cross-validation. Lambda was determined when the partial likelihood deviance was smallest (Right). (c) The risk score of the different risk groups. (d) The survival probability of the different risk groups. Blue represents high-risk patients and red represents low-risk. The patients were divided according to their median risk score. (e) Prognostic genes in the TCGA-HCC dataset were analyzed using ROC curves.
