Fig. 3

The construction and evaluation of the risk model in the training set. (a) Univariate Cox regression analysis of 26 DECRGs to identify prognostic genes. (b) The error plot of LASSO cross-validation. The dashed line on the left represents lambda.min, which is the position of the minimum cross-validation error, and the number of feature genes is shown above it. The dashed line on the right represents the line with fewer features. (c) The LASSO coefficient path plot. The lines in different colors correspond to different genes and their coefficients; as lambda increases, the coefficients of the feature variables approach zero, which is the optimal condition for selecting the target genes. (d-e) Risk score distribution and associated survival status of samples in the training set. (f) KM survival curve comparing overall survival between the high-risk and low-risk groups. (g) ROC curve evaluating the predictive performance of the prognostic model in the training set. Abbreviations: DECRGs: differentially expressed circadian rhythm-related genes; LASSO: least absolute shrinkage and selection operator; KM: Kaplan–Meier; ROC: receiver-operating characteristic