Bioinformatics analysis of immune-related differentially expressed genes in pediatric Kawasaki disease

Abstract Kawasaki disease (KD) is an acute pediatric systemic vasculitis of unknown origin, where emerging evidence suggests that immune dysregulation and alterations in peripheral immune cell populations play critical roles in its pathogenesis. In this study, we performed comprehensive bioinformatics analyses to identify KD-associated genes and develop an accurate diagnostic model for early detection. Two publicly available microarray datasets, GSE73461 (132 samples) and GSE68004 (126 samples), were obtained from the Gene Expression Omnibus (GEO) database and analyzed.Immune-related differentially expressed genes (IMRDEGs) were identified by integrating differential expression profiling with Gene Ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. To select optimal biomarkers, we employed a multi-algorithm framework combining logistic regression, SVM-RFE, and LASSO regression. Among 12 IMRDEGs identified, five genes (ITGAM, CAMP, CD4, IL2RB, and IL1B) demonstrated the highest diagnostic potential, with ITGAM showing the strongest predictive value. Functional enrichment revealed these genes’ involvement in leukocyte activation and immune-related signaling pathways, including tuberculosis signaling.Comparative immune profiling highlighted significant differences in 14 immune cell populations between KD patients and healthy controls. Overall, our findings uncover critical immune-related genes linked to KD pathogenesis and propose a robust diagnostic model with promising clinical applicability.