Bioinformatic Analysis And Machine Learning To Identify The Biomarkersand Immune Infiltration Of Atopic Dermatitis

Atopic dermatitis(AD)is a common inflammatory skin disease that has both genetic and environmental factors in its etiology.The disease presents with chronic,recurrent,and long-term features leading to an impact on the patient’s quality of life.To date,numerous studies have been performed on AD.However,an effective diagnostic test has yet to be identified.To date,researchers have conducted a great deal of research on AD.However,no effective diagnostic method has been established.Therefore,it is of great significance for the treatment and prevention of AD patients to search for molecular diagnostic targets of AD together with bioinformatics and almethods and establish diagnostic methods.Methods:This article first conducts differential expression analysis of AD gene expression profiling data and uses enrichment analysis to annotate the functions of differentially expressed genes.Subsequently,weighted correlation network analysis(WGCNA)is used to screen for genes associated with AD skin lesions.The intersection genes of differentially expressed genes and key modules are selected for further feature gene screening using random forest(RF),least absolute shrinkage and selection operator(LASSO),and support vector machine-recursive feature elimination(SVM-RFE)analysis.Finally,the immune cell infiltration in atopic dermatitis is evaluated and the correlation between feature genes and immune infiltrating cells is analyzed.Results:A total of 345 differentially expressed genes were screened,including 136 up-regulated genes and 209 down-regulated genes.Gene function enrichment was associated with adaptive immune response,cell chemotaxis,chromosome segregation,antiviral defense response,granulocyte chemotaxis,and so on.Pathway enrichment analysis showed that the cell cycle,chemokine signaling pathway,cytokine-cytokine receptor interaction,JAK-STAT signaling pathway,primary immunodeficiency,and other signaling pathways were significantly enriched in AD skin lesion samples.Using machine learning,four genes were finally screened:WNT5A,S100A7A,ACADL and WIF1.Immune infiltration analysis showed that resting mast cells,helper T cells,CD4+ memory T cells,naive CD4~+T cells,dendritic cells,activated NK cells,B memory cells,and macrophages were involved in the pathogenesis of atopic dermatitis.Conclusion:WNT5A,S100A7A,ACADL,and WIF1 may serve as diagnostic biomarkers for atopic dermatitis,and the occurrence of AD is associated with abnormal immune reactions.