Fig. 3

Construction of a 6 SE-lncRNAs risk model. A Cross-validation for adjusting parameter choice in LASSO regression analysis. B 6 SE-lncRNAs and the corresponding coefficients were shown. C Kaplan–Meier survival curves of two risk groups in the CGGA dataset. D The ROC curve indicates the sensitivity and specificity of the risk signature to predict prognosis in the training set. E The heatmap displayed the discrepant distribution of clinical features and expression of 6 SE-lncRNAs in low- and high-risk glioma patients. F–M The WHO grade (F), cluster 1/2/3 subgroups (G), age (H), fustat (I), MGMT promoter methylation status (J), 1p/19q codel status (K), PRS type (L), and IDH status (M) stratify the training set, and the distribution of risk scores was shown. N Surveying the risk model in CGGA dataset: distribution and survival status of glioma patients in low- and high-risk groups. O, P Univariate (O) and multivariate (P) analyses for the training set consisting of gender, age, grade, and risk score. *p < 0.05, **p < 0.01, ***p < 0.001