Fig. 4

Eleven-genes-based prognostic model construction. a Lasso-Cox regression coefficient selection and variable screening. b Cross-validation in the LASSO-Cox regression model to select the tuning parameter. c. Display of regression coefficients corresponding to filtered variables. d. Heat maps of gene expression in prediction models. e. Validation of KM curves for models. f. Validation of ROC curves. g. Distribution of risk scores for all samples. h. Scatter plots of survival time for all patients. i. Correlation of risk scores with various clinical characteristics. j. Analysis of the difference in the proportion of immune-infiltrating cells between high- and low-risk groups. k Prediction of chemotherapy resistance in patients from high and low-risk groups. * p < 0.05, ** p < 0.01, and *** p < 0.001