Research Paper Volume 16, Issue 5 pp 4518—4540
Tumor-infiltrating macrophage associated lncRNA signature in cutaneous melanoma: implications for diagnosis, prognosis, and immunotherapy
Figure 2. Prognostic macrophage associated differentially expressed lncRNAs (DEMlncRNAs) feature selection. (A) Forest plots of the prognostic DEMlncRNAs in training dataset. (B) Misclassification error distribution with the corresponding λ-logarithm value and the left variants of model in The Least Absolute Shrinkage and Selection Operator (LASSO) algorithm. (C) Random forests variable selection (RF-VS) algorithm. The lowest point of the curve indicates the lowest out of bag (OOB) error, and the corresponding DEMlncRNAs at this point are the best signature selected by RF-VS. (D) Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithms. The point highlighted indicates the lowest error rate, and the corresponding DEMlncRNAs at this point are the best signature selected by SVM-RFE. (E) The Venn plot of overlapping DEMlncRNAs selected by LASSO, RF-VS and SVM-RFE algorithms. (F) Boxplots of lncRNA expression distribution in GSE15605 dataset.