Fig. 8: Construction of a ccRCC prognostic risk model based on ccRCC-specific enhancer-hijacking events using machine learning.
From: Structural variation drives enhancer hijacking via 3D genome disruption in ccRCC
Receiver Operating Characteristic (ROC) curves demonstrating the robust prognostic predictive performance of the risk model in the TCGA-KIRC cohort (a), training cohort (b), and testing cohort (c). Time-dependent Receiver Operating Characteristic (ROC) curves evaluating the risk model’s performance at 1-, 3-, and 5-year intervals in the TCGA-KIRC cohort (d), training cohort (e), and testing cohort (f). g Calibration curve analysis validating the stability and reliability of model predictions. h Nomogram for predicting 1-, 3-, and 5-year overall survival (OS) in ccRCC patients within the TCGA-KIRC cohort. Kaplan-Meier survival curves depicting significant divergence in overall survival (OS) (i) and progression-free survival (PFS) (j) between high- and low-risk groups in the TCGA-KIRC cohort. Kaplan-Meier survival analysis showing distinct overall survival (OS) (k) and progression-free survival (PFS) (l) outcomes for high- versus low-risk groups in the training cohort. Kaplan-Meier survival curves confirming differential overall survival (OS) (m) and progression-free survival (PFS) (n) between risk strata in the testing cohort.
