Abstract

Objective: Childhood allergic asthma (CAA) is a chronic allergen-driven airway inflammatory disease whose molecular mechanisms and precise diagnostic and therapeutic strategies remain incompletely understood. This study aimed to identify mitochondrial–mitophagy-related genes (MMRGs) with diagnostic and therapeutic potential in CAA.Methods: Public transcriptomic datasets were analyzed using weighted gene co-expression network analysis and multiple machine learning algorithms to identify diagnostic signature genes and construct a CAA diagnostic model. Immune infiltration, competing endogenous RNA network, and drug–gene interaction analyses were subsequently performed. The expression and biological functions of selected genes were further evaluated in an IL-13-stimulated airway epithelial cell model.Results: Fifteen MMRG-related signature genes with diagnostic potential were identified, and the resulting diagnostic model exhibited good predictive performance. Several immune cell subsets, including resting mast cells, and naive B cells, showed altered infiltration in CAA, and their abundances were significantly correlated with signature-gene expression. The constructed ceRNA network indicated complex post-transcriptional regulation of these genes. Drug–gene interaction analysis identified multiple potential therapeutic compounds, with BCL2-related agents being the most abundant. In IL-13-stimulated airway epithelial cells, MRPL43 expression was increased, whereas BCL2 expression was decreased. MRPL43 knockdown significantly restored cell viability and reduced the secretion of pro-inflammatory cytokines.Conclusions: MMRGs are closely associated with the diagnosis, immune microenvironment, and potential therapeutic regulation of CAA. MRPL43 may contribute to airway epithelial inflammation and represents a potential biomarker and therapeutic target for CAA.