Figure 4. Construction of scoring system and analysis of its prognostic value in the TCGA cohort. (A) LASSO coefficient profiles of 192 FARGs. (B) Selection of the penalty parameter (λ) in the LASSO model via 1,000 cross-validations. The vertical dotted line passing through the red dot is drawn at the optimal value. (C) Nine FARGs screened by multi-Cox regression analysis were used to construct the scoring system. (D, E) The K-M survival curves showed that the OS and PFS of glioma patients in the low-risk group were longer than those in the high-risk group. (F, G) ROC curves show that risk score can effectively predict the 1,3,5-year OS and PFS for glioma patients. (H) The distributions of five clinicopathological characteristics and nine FARGs between high and low-risk groups. (I, J) Uni- and multi-Cox regression analysis of risk score and five clinicopathological characteristics. *P<0.05, **P<0.01, ***P<0.001.