Identification of Mitochondrial-Related Signature in Allergic Rhinitis: Novel Insights from Bioinformatics and Machine Learning.

BACKGROUND: Allergic rhinitis (AR) is an immune-mediated inflammatory disease of the nasal mucosa. Numerous studies have demonstrated correlations between AR and mitochondrial dysfunction. The present study was designed to explore mitochondrial-related genes associated with AR. METHODS: The differentially expressed mitochondrial-related genes (DEMRGs) were obtained by intersecting differentially expressed genes (DEGs) and mitochondrial-related genes. Further screening of hub DEMRGs was conducted using three machine learning algorithms. The CIBERSORT algorithm was employed to characterize immune cell infiltration and investigate interactions between the hub DEMRGs and immune cells. Ultimately, the expression levels of hub DEMRGs were confirmed in external datasets and clinical data from human nasal mucosa. RESULTS: In total, 12 DEMRGs were identified, and using machine-learning algorithms, two hub DEMRGs (SLC25A36 and RNASEH1) were significantly downregulated in AR samples, indicating good diagnostic potential. These hub DEMRGs were closely associated with the infiltration of activated CD4+ memory T cells, resting dendritic cells, M0 macrophages, and resting NK cells. Validation using external datasets and experiments confirmed that the two hub DEMRGs were downregulated in the AR group, consistent with the bioinformatics analysis. CONCLUSION: By characterizing both mitochondrial-related genes and patterns of immune cell infiltration in AR, this study suggested a potential link to AR pathogenesis and identified candidate targets for future therapeutic exploration.