Abstract
BACKGROUND: Atherosclerosis is a chronic inflammatory disease involving mitophagy and immune dysregulation. Currently, there is no integrated diagnostic model for autophagy immune genes in mitochondria. This study aimed to identify potential diagnostic markers through integrated bioinformatics and experimental validation. METHODS: Two atherosclerosis-related datasets (GSE43292 and GSE100927) were analyzed to identify differentially expressed mitophagy-related genes (MRGs), followed by enrichment analysis. Key genes were screened using LASSO, SVM-RFE, and random forest algorithms. A diagnostic nomogram was constructed and validated by ROC analysis. Immune infiltration was evaluated using CIBERSORT and ssGSEA. GSEA, GSVA, and unsupervised clustering were applied to explore biological pathways and molecular subtypes. qPCR validation was performed in ox-LDL-treated RAW264.7 and THP-1 cells. RESULTS: Thirteen upregulated and six downregulated MRGs were identified. Five hub genes (MNDA, CD163L1, NEXN, TC2N, SLC22A3) demonstrated strong diagnostic performance (AUC > 0.85) and were closely associated with immune cell infiltration; two molecular subtypes with distinct immune profiles were identified; qPCR validation confirmed the differential expression of these genes under inflammatory stimulation. CONCLUSION: MNDA, CD163L1, NEXN, TC2N, and SLC22A3 may serve as diagnostic biomarkers for atherosclerosis. This five-gene model can stratify patient risk and guide personalized anti-inflammatory/autophagic therapy.