Identification Of Biomarkers For Early Diagnosis Of Frozen Shoulder By Weighted Gene Co-Expression Network Analysis And Their Clinical Relevance

Background: Frozen shoulder,also known as adhesive capsulitis,etc.,is a chronic inflammatory disease with limited glenohumeral joint motion and pain as the main clinical manifestations,which seriously affects patients’ quality of life,and its pathogenesis is not fully understood.The aim of this study was to analyze the differential genes between the focal tissues of patients with frozen shoulder and those with shoulder instability by bioinformatics methods to provide a theoretical basis for molecular biological studies and biomarker screening of frozen shoulder.Methods: The GSE140731 dataset was found from the GEO and Array Express databases.High-throughput sequencing data were extracted from tissue samples of 22 patients with frozen shoulder and 26 control patients with shoulder instability,and differentially expressed genes were analyzed using R language and Deseq2 package,which led to Gene ontology(GO)and genome encyclopedia of genes and genomes(KEGG)pathway enrichment and GSEA analysis of key pathways enriched for frozen shoulder,combined with weighted gene co-expression network analysis(WGCNA)and lasso regression(LASSO),to find key genes that play a role in the disease.Subsequently,diagnostic values were assessed using subject operating characteristic curves(ROC)and a single sample GSEA was used to examine the level of infiltration of 28 immune cells and to assess the relationship between relevant immune cells and diagnostic markers.Results: A total of 387 differentially expressed genes were screened,including 342 upregulated genes and 45 down-regulated genes,and functional and pathway enrichment analysis of the differential genes was performed,and four co-expression modules were obtained by WGCNA;among them,one hub module(turquoise module)had the highest correlation with frozen shoulder.Six pivotal genes ADAMTS4,COL11A1,MARCKSL1,MFAP1,SCG2 and SPON2 associated with frozen shoulder were further obtained,and all the above core genes were expressed up-regulated in adhesive capsulitis tissues.ROC curve analysis indicated that the six pivotal genes had major diagnostic value.Conclusion: With the help of WGCNA and LASSO,combined with bioinformatics analysis of ss GSEA,and by integrating multiple bioinformatics tools,we screened a pivotal module(turquoise module)and identified six pivotal genes(ADAMTS4,COL11A1,MARCKSL1,MFAP1,SCG2 and SPON2),which may become new markers and immunotherapeutic targets.