Integrative transcriptomic and machine learning analysis identifies PYCARD and IFI30 as immune-lysosomal biomarkers of ANCA-associated glomerulonephritis.

OBJECTIVES: ANCA-associated glomerulonephritis (ANCA-GN) is an immune-mediated kidney disease leading to acute or chronic renal failure. This study investigates the role of mitophagy-related genes in ANCA-GN, as mitochondrial dysfunction is closely linked to the pathogenesis of various kidney diseases. METHODS: This study analyzed transcriptomic data from GEO datasets (GSE104948 and GSE108109) to investigate mitophagy-related mechanisms in ANCA-GN. Methods included batch correction, consensus clustering (identifying two subtypes), weighted gene co-expression network analysis (WGCNA), differential expression screening, and machine learning (LASSO, random forest, SVM-RFE). A diagnostic nomogram was constructed and validated, and immune cell infiltration was profiled. RESULTS: Analyses revealed distinct activation of immune pathways, including complement and phagosome signaling, alongside abnormal infiltration of CD8+ T cells in ANCA-GN. Subtype-specific analysis identified 131 differentially expressed genes (DEGs), while 143 DEGs distinguished ANCA-GN from controls.Intersection analysis and machine learning prioritized two hub genes, PYCARD and IFI30, which exhibited strong diagnostic accuracy (AUC >0.9) and correlated with CD8+ T-cell infiltration. A nomogram model validated their clinical utility (AUC >0.9). Functional enrichment highlighted phagocytosis and immune signaling pathways. Immune profiling revealed significant upregulation of 20 immune cell types in ANCA-GN. CONCLUSIONS: These findings suggest that mitophagy-immune crosstalk drives ANCA-GN progression, with PYCARD and IFI30 as potential diagnostic biomarkers. This study provides mechanistic insights into ANCA-GN pathogenesis and proposes novel targets for clinical intervention.