Screening Of Core Genes And Drug Prediction In Dermatomyositis Based On GEO Database

Objective Dermatomyositis is a multifactorial chronic autoimmune disease that often involves multiple organs throughout the body and is associated with a high frequency of tumors.Due to the rarity of the disease,its etiology and pathogenesis have not been fully elucidated,related complications have poor prognosis,and treatment is limited to traditional regimens such as hormones and immunosuppressants.Therefore,it is particularly urgent to find new treatment methods.Based on the method of bioinformatics,this study analyzes the microarray data in the public database of gene expression profiles that meet the screening criteria,and deeply explores the genetic situation of patients with DM and its possible molecular biological functions,with a view to finding potential biomarkers that affect the occurrence and development of DM,thereby mining small molecular chemical drugs and Chinese herbal medicines for the treatment of DM,Provide theoretical reference and ideas for clinical precision treatment and new drug development.Methods We searched the Gene Expression Omnibus(GEO)database with "Dermatomyositis" as the search term,downloaded the eligible microarray data sets,and used three sets of muscle tissue chip data as the training set and one set of skin tissue chip data as the validation set.Based on the integration of the three muscle tissue microarray datasets,the differentially expressed genes were analyzed using R language,followed by GO and KEGG enrichment analysis of differentially expressed genes to obtain the pathway profiles of DM.Protein interaction networks were constructed using the STRING database and visualized in Cytoscape software.The most important modules were identified by the MCODE plugin in Cytoscape,and core genes were filtered using the Degree algorithm of the Cyto Hubba plugin.The training set in the microarray dataset was subjected to ROC curve analysis and the validation set was verified by statistical analysis to validate the accuracy of these core genes as biomarkers.The percentage of immune cells was obtained after CIBERSORT immune infiltrating cell analysis for differentially expressed genes,and the correlation between the expression of the identified core genes and the abundance of immune infiltrating cells was investigated using Spearman correlation analysis.Finally,the validated core genes were predicted and screened for small molecule chemical drugs and Chinese herbal medicines,respectively.Results A total of 552 differentially expressed genes were found in the muscle tissue of DM compared with healthy controls,of which 402 were up-regulated and 150 were down-regulated in expression.By GO enrichment analysis,the results were mainly enriched in cytoplasm,cytosol,plasma membrane,nucleus and extracellular exosome,functionally bound to protein and identical protein binding,protein homodimerization activity,calcium ion and zinc ion binding,mainly involved in signal transduction,innate immune response,defense response to virus,immune response,positive regulation of transcription from RNA polymerase II promoter and other biological processes.The up-regulated genes enrichment pathways are mainly concentrated in Epstein-Barr virus infection,influenza-A,cytokine-cytokine receptor interactions and other pathways,while the down-regulated genes are mainly enriched in calcium,glucagon signaling pathways and chemical carcinogenesis-receptor activation.The PPI network constructed by target genes were screened by degree algorithm to identify STAT1,HLA-A,IFIT1,ISG15,MX1,RSAD2,IFIT3,HLA-C,IRF7,HLA-F as key genes.The ROC curve areas of these 10 genes were above 0.75,and all of them were statistically different in the statistical analysis of the validation set.Under CIBERSORT algorithm analysis,plasma cells,M2 and M1 macrophages were significantly and significantly higher in dermatomyositis muscle tissue than in control group.The validated core genes were strongly positively correlated with MI macrophages,plasma cells,M2 macrophages,and NK cells activated,and significantly negatively correlated with T cells regulatory,mast cells activated,and CD8+ T cells.By analyzing the core genes,the top 10 small molecule chemical drugs were predicted as catechin,naringin,mitotane,alitretinoin,cilnidipine,tacedinaline,erythromycin,amsacrine,bergenin,entinostat,respectively,and the Chinese herbs related to disease were selected as Leigengteng,Danshen,Sanqiye,Sanqihua,Sanqi,Chashugen,Yuancane,Cansha,Houpuhua,Houpu.Conclusion STAT1,HLA-A,IFIT1,ISG15,MX1,RSAD2,IFIT3,HLA-C,IRF7,HLA-F 10 core genes,plasma cells,macrophages and viral immunity and other related pathways can be used as potential biomarkers and molecular mechanisms of DM,and 10 small molecule drugs and traditional Chinese medicines may have potential therapeutic effects,providing theoretical basis and ideas for precise clinical treatment of DM and the development of new drugs.