Integrated Application of Bioinformatics and Machine Learning to Investigate Apoptosis-Related Genes as Diagnostic Biomarkers and Develop Predictive Models in the Pathogenesis of Behçet's Disease and Atherosclerosis

This study investigates apoptosis-related mechanisms in Behçet's disease (BD)-associated atherosclerosis (AS) through integrated bioinformatics and machine learning. Analysis of datasets (GSE209567, GSE100927) identified BD-specific differentially expressed genes (DEGs) and AS-related hub genes via limma and WGCNA. Intersection with apoptosis-related genes revealed BD-AS-ARGs. Functional enrichment and PPI network analyses highlighted key pathways, while machine learning (LASSO, RF, GBM, XGBoost) prioritized four hub genes (MMP9, SPI1, FOXO3, ITGB3). A diagnostic nomogram demonstrated high accuracy (AUC: 0.986, 95% CI: 0.969–1.000), suggesting these genes as biomarkers and therapeutic targets for BD-induced AS pathogenesis.