Shared Diagnostic Genes and Potential Mechanism between Allergic Rhinitis and Atopic Dermatitis revealed by Integrated Transcriptomic Analysis and Machine Learning

Xiaojing Zhang¹Meng Sun²Liang Xu³Junjie Bi²Xiangjing Chen²Lei Yao²Yuan Jia²Ying Zhang¹Wei Meng^2*

• ¹Shandong University of Traditional Chinese Medicine, Jinan, China
• ²Shandong University of Traditional Chinese Medicine Affiliated Hospital, Jinan, China
• ³Central Hospital Affiliated to Shandong First Medical University, Jinan, China

Allergic rhinitis and atopic dermatitis are among the most prevalent chronic allergic diseases, often co-occurring and sharing overlapping immunopathological features. However, the molecular mechanisms underlying their comorbidity remain incompletely understood. In this study, we conducted an integrative bioinformatics analysis to identify shared diagnostic biomarkers and potential mechanistic connections between AR and AD. By analyzing transcriptomic datasets, we identified 36 overlapping differentially expressed genes. A combination of protein to protein interaction network analysis and random forest modeling revealed five hub genes: CD274, CYP2E1, FOLH1, SERPINB4, and SPRR1B, all showing strong diagnostic value in both diseases. Functional enrichment analysis indicated their involvement in epithelial barrier regulation, immune cell signaling, and oxidative stress pathways. Immune infiltration profiling demonstrated that these genes were significantly associated with dendritic cells, T cells, and natural killer cells in both AR and AD cohorts. In addition, miRNA to mRNA interaction networks and drug to gene associations further highlighted their translational relevance. These findings suggest that AR and AD share a common molecular landscape and nominate robust biomarkers for early diagnosis and therapeutic targeting in allergic conditions.