Fig. 2

A, B Screening the 12 most representative genes expressed in response to TEC genes by LASSO-Cox analysis. C Histogram describing the important features of the LASSO-Cox predictive model for OS. D, E Relationship between the error rate and number of classification trees. F Histogram describing the important features of the random forest predictive model for OS. G The negative partial log-likelihood of the Cox proportional hazards regression in the XGBoost model changes as the number of iterations increases. H Histogram describing the important features of the XGBoost predictive model for OS. I Intersection of the genes identified by the three algorithms