Identification and Exploration of Novel B Cell Infiltration-Related Biomarkers in Endometriosis

OBJECTIVE: To explore B cell infiltration-related genes in endometriosis (EM) and investigate their potential as diagnostic biomarkers. METHODS: Gene expression data from the GSE51981 dataset, containing 77 endometriosis and 34 control samples, were analyzed to detect differentially expressed genes (DEGs). The xCell algorithm was applied to estimate the infiltration levels of 64 immune and stromal cell types, focusing on B cells and naive B cells. Weighted gene coexpression network analysis (WGCNA) identified B cell infiltration-related gene modules. Potential biomarker genes were screened using LASSO and SVM-RFE machine learning methods. Then, hub genes were validated in an independent GSE7305 dataset, and Pearson correlation analysis was used to assess associations between hub genes and B cell markers. RESULTS: A total of 4341 DEGs were screened and greenyellow module containing 349 genes were associated with infiltration characteristics of B cells in EM lesions, then 12 B cell infiltration-related genes were identified by machine learning methods. Based on the external GSE7305 dataset, four hub genes, of which NR4A1, TNS1, ZNF521, and CMPK2, were recognized as potential biomarkers of B cell infiltration in EM, and all of them were significantly upregulated. CONCLUSION: This study identified and exploration four potential diagnostic biomarkers in EM. The functions of the four biomarkers and the role of B cell infiltration in EM were determined using bioinformatics analysis, providing new insights into endometriosis at the immune and molecular levels.