Figure 5.

(A) Assessment of the accuracy of the model in predicting the prognosis of prostate cancer patients in the training cohort using ROC curves; and (B) The same methodology to assess the test cohort in the test cohort. (C) Distinguishing patients in the training set by PCA results of the DRGs risk model; (D) Applying the same model to distinguish patients in the test set; (E) Analyzing PFS survival curves of different risk groups in the external validation cohort GSE46602; and (F) Demonstrating heatmaps of the correlation between clinical characteristics and risk scores.