Identification of two potential biomarkers of Graves' Disease by bioinformatics analyses based on Gene Expression Omnibus datasets with immune cells infiltration.

Abstract Background: Graves' disease (GD) is an autoimmune disease, the incidence of which is increasing year by year. And it still needs long life therapy. So, the detailed pathogenesis of GD still needs further study. Method: In our study, differentially expressed genes (DEGs) were derived from online Gene Expression Omnibus (GEO) microarray expression dataset GSE71956. Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes (KEGG) analysis was used to enrichment analysis. Protein-protein interaction (PPI) network analyses were used to identify the hub genes. Then hub gene was validated by qPCR. Next CIBERSORT analysis was used to further explore the immune infiltration among hub genes. ROC curve was used to analysis the specificity and sensitivity of hub genes. Result: 44 DEGs were screened out from GEO dataset. GO and KEGG analysis of DEGs shows that GD involves immunity and metabolism. Two hub genes EEF1A1, EIF4B were recruited from PPI network, and was validated by qPCR ( p < 0.05). Immune cell infiltration analysis revealed that plasma cells, T cells CD4 memory resting, T cells follicular helper, dendritic cells activated, mast cells activated were correlated with GD. We are the first to demonstrate the expression of EEF1A1, EIF4B has significant positive correlations with infiltrating levels of T cells CD4 memory resting. The AUC of EEF1A1 ROC analysis is 0.687 and ROC of EIF4B is 0.733.Conclusion : Our study revels two hub genes, EEF1A1, EIF4B, which were associated with resting memory CD4 + T cells , may be the potential molecular biomarkers and therapeutic targets of GD.