Fig. 5 | Scientific Reports

Fig. 5

From: Identification of a novel disulfidptosis-related gene signature in osteoarthritis using bioinformatics analysis and experimental validation

Fig. 5

Screening of key DR-DEGs using multiple machine learning algorithms. (A) Random forest (RF) model. (B) Ranking the importance of the 10 DR-DEGs based on MeanDecreaseGini values. (C) ROC curves for RF, SVM, and GLM. (D) Importance ranking of feature genes in RF, SVM, and GLM. (E) Reverse cumulative distribution of residual in RF, SVM, and GLM. (F) Box plot of residual in RF, SVM, and GLM. (G) Coefficient distribution plot of Lasso regression analysis. (H) Cross-validation curve. (I) Intersection of DR-DEGs from four machine learning algorithms. (J) Correlation analysis of GYS1 and LRPPRC. Single-gene GSEA in SLC3A2 (K), GYS1 (L), LRPPRC (M) and PPM1F (N).

Back to article page