Identification and validation of metabolism-related genes signature and immune infiltration landscape of rheumatoid arthritis based on machine learning

Zhaoyang Guo; Yuanye Ma; Yaqing Wang; Hongfei Xiang; Huifei Cui; Zuoran Fan; Youfu Zhu; Dongming Xing; Bohua Chen; Hao Tao; Zhu Guo; Xiaolin Wu

doi:doi:10.18632/aging.204714

← Back

Research Paper Volume 15, Issue 9 pp 3807—3825

Identification and validation of metabolism-related genes signature and immune infiltration landscape of rheumatoid arthritis based on machine learning

Figure 5. ROC curve analysis and the expression of the diagnostic feature biomarkers. The violin diagram shows the expression of (A) AKR1C3, (B) MCEE, (C) POLE4, and (D) PFKM. (E) A nomogram model to validate the impact of the four feature biomarkers on diagnostic effectiveness. ROC curve analysis of (F) MCEE, (G) AKR1C3, (H) POLE4 and (I) PFKM. (J) The AUC value of the nomogram score was 92.8%. *P < 0.05; **P < 0.01; ***P < 0.001; **** P < 0.0001.