Journal: Frontiers in Genetics
Article Title: Integrative analysis of single-cell and microarray data reveals SPI1-centered macrophage regulatory signatures in ulcerative colitis
doi: 10.3389/fgene.2025.1617834
Figure Lengend Snippet: Identification of SPI1-regulated hub genes through the integration of gene set intersection and machine learning approaches. (A,B) LASSO regression analysis with cross-validation curve showing the optimal lambda value selection and coefficient profiles for gene selection. (C) Accuracy versus number of variables selected using RFE. (D) Bar chart of top genes ranked by importance scores. (E) SVM-RFE identified a 30-gene signature that achieved the highest classification accuracy. (F) Venn diagram illustrating the overlap between genes selected by three machine learning methods (LASSO, RFE-RF, and SVM-RFE). (G,H) Expression levels of SPI1, IRAK3, IL1RN, CD55, and PEA15 in two independent datasets GSE87466 (G) and GSE75214 (H) . (I) ROC curves for IRAK3, IL1RN, CD55 and PEA15 in the training dataset. (J) ROC curves for IRAK3, IL1RN, CD55 and PEA15 in the validation dataset. **** P < 0.0001 compared to the control group.
Article Snippet: ELISA kit for IL-1β (ZC-37974, Zhuocai, China); ELISA kit for IL-10 (ZC-37962, Zhuocai, China); SPI1 antibody (55100-1-AP, Proteintech, China); iNOS antibody (22226-1-AP, Proteintech, China); Arg1 antibody (16001-1-AP, Proteintech, China); Tubulin antibody (80762-1-RR, Proteintech, China); Anti-rabbit IgG (H + L) (14780, Cell Signaling Technology, United States).
Techniques: Biomarker Discovery, Selection, Expressing, Control