Fig. 5

Construction and validation of fibrogenesis-related prognostic model. A, B 6 candidate pFRGs identified by LASSO regression with ten-fold cross-validation, using the minimum partial likelihood deviance to determine the optimal lambda. C Coefficient profiles of 6 optimal pFRGs selected for the model. D-F ROC curves demonstrating the predictive performance of the prognostic model in the training (D), test (E) and whole (F) sets. G K–M survival showing a different prognosis in high- and low-risk groups. H, I Model performance validation using an external GEO dataset with ROC (H) and K-M (I) curves. J Risk plot of survival status of ccRCC patients and expression profiles of 6 pFRGs between risk groups. K A nomogram for estimation of survival rates for ccRCC patients. L Calibration plots of the nomogram at 1, 3, and 5 years. AUC area under curve, HR hazards ratio