Research Paper Volume 16, Issue 6 pp 5149—5162
A novel immune-related lncRNA signature predicts the prognosis and immune landscape in ccRCC
Figure 3. Construct an optimal risk assessment model. (A) Cross-validation for tuning parameter selection by the modified least absolute shrinkage and selection operator analysis (LASSO). (B) LASSO coefficient profiles of 38 prognostic immune-related genes. (C) The univariable Cox model result was shown by forest plot. (D) Calculate the maximum AIC value generated by ROC curves of 38 DEirlncRNAs and get the optimal cut-off value. (E) The 1-, 3-, 5-year ROC of the optimal model revealed that all AUC values were over 0.90. (F) A comparison of 5-year ROC curves with other common clinical characteristics showed the superiority of the riskScore.