Integrating bioinformatics and machine learning to analyze differentially expressed genes in atopic dermatitis and their relation-ship with immune cell infiltration
Objective By utilizing bioinformatics and machine learning strategies,to preliminarily explore differentially expressed genes(DEGs)and the relationship with immune cell infiltration in atopic dermatitis(AD)to enhance understanding and diagnosis of the disease.Methods The study obtained two gene expression datasets by retrieving the Gene Expression Omnibus(GEO)database,including normal skin of 20 healthy volunteers and affected skin of 51 untreated AD patients used for modeling,and the normal skin of another 22 healthy volunteers and affected skin of 30 AD patients used for validation,in order to identify DEGs.The DEGs were validated using the least absolute shrinkage and selection operator regression model and support vector machine-recursive feature elimination.Inflammation status of AD patients was assessed using CIBERSORT to further explore the correlation between DEGs and immune cells.Results Compared to normal subjects,AD patients exhibited differential expression of MAPK-activated protein kinase 5 antisense RNA 1(MAPKAPK5-AS1).Additionally,significant differences were observed in the proportions of six types of immune cells between the skin of AD and normal control groups.The expression level of MAPKAPK5-AS1 was positively correlated with the numbers of naive B cells,activated CD4+memory T cells,and activated mast cells.Conclusion MAPKAPK5-AS1 may be an important DEGs in AD,which is associated with immune cell infiltration characteristics,providing new insights for the diagnosis and treatment of AD.
国家科技期刊平台
NSTL
万方数据
