In-silico identification of differentially expressed genes in Type 1 diabetes mellitus

Abstract

The incidence of Type 1 Diabetes Mellitus T1DM varies markedly in different geographical populations but seems to be increasing globally. The focus of this research is to screen for T1D-associated differentially expressed genes (DEGs). A meta-analysis was conducted using the Gene Expression Omnibus (GEO) datasets. The datasets included samples from T1DM and normal patients. The Robust Multichip Averaging (RMA) procedure was used for background correction, normalization and summarization to obtain expression level data and to discover differentially expressed genes. Box plots, Density plots, RNA degradation plots and recommended procedures from Affymetrix for quality control were implemented. The DEGs were screened and the exclusively expressed genes were uncovered through the Venn diagrams and heat maps functions in R language. 3,824 genes were classified, as DEGs of which 2,030 were upregulated and 1,794 were downregulated. Seven key genes (TLN1, ANPEP, F13A1, SPARC, SPTBN1, IGHA2 and IGHA1) were exclusively expressed in the whole progression. 58 DEGs were revealed through the Venn diagrams while the Heatmaps showed the differential expression data for 35 genes. IGHA1, IGHA2, IGKV4-1 were significantly expressed and upregulated. Although some of these genes have been previously associated with T1D, many other genes were identified for further studies.