Machine learning–driven identification and experimental validation of key biomarkers in the bile acid metabolic pathway associated with ulcerative colitis

Le résumé fourni par la source

Background: Bile acids are shown to participate in inflammatory responses. This study was designed to investigate the functions of bile acid metabolism-associated genes (BAMGs) in ulcerative colitis (UC), identify the potential biomarkers based on eleven machine learning algorithms. Methods: Seven independent UC transcriptomic datasets were retrieved from the GEO database. Differentially expressed genes, weighted gene co-expression network analysis (WGCNA), and multiple machine learning algorithms were integrated to identify key BAMGs. Subsequently, enrichment analysis, immune cell analysis and single cell analysis were performed to explore the biological functions and immunological characteristics. The dextran sulfate sodium (DSS) induced colitis model in mice was then established and validated the results through western blot and immunohistochemical (IHC) analysis. In addition, peripheral blood samples were collected from UC patients for the detection of feature gene expression by quantitative real-time PCR (RT-qPCR). Results: ) were identified. Unsupervised clustering based on the three-gene signature stratified UC patients into two distinct subgroups exhibiting divergent immune status. In DSS-treated mice, western blot and IHC confirmed significantly reduced SLC23A1 and PHYH protein levels and elevated CH25H protein expression in colonic tissues. RT-qPCR analysis of PBMCs from UC patients showed consistent gene expression. Immune cell analysis showed obvious association between the key BAMGs and inflammatory cells including naïve B cells, neutrophils, monocytes, CD8 T cells, and macrophages. Single-cell analysis revealed that the three feature genes were differentially expressed across T- and B-cell subsets, indicating their potential involvement in UC. Conclusion: may contribute to UC pathogenesis and represent potential biomarkers.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.