Background
Rheumatoid arthritis (RA) is a type of systemic immune disease characterized by chronic inflammatory disease of the joints. However, the etiology and underlying molecular events of RA are unclear. Here, we applied bioinformatics analysis to identify the potential biomarkers involved in RA.
Methods
The three microarray datasets (GSE1919、GSE10500 and GSE55457) were downloaded from the Gene Expression Omnibus (GEO) database. We used the R software screen the differentially expressed genes (DEGs). These DEGs shared by the three microarray datasets were further identified. Next, we carried out functional enrichment analysis using the Gene Ontology (GO) and Kyoto Encyclopaedia of Genes and Genomes (KEGG). Then, the hub genes with a relatively high number of connections to other DEGs, were identified by Cytoscape. Other bioinformatics methods are also performed, including protein–protein interaction (PPI) network analysis and construction of miRNA–hub gene networks and transcription factors (TF)–hub gene networks.
Results
A total of 115 overlapping DEGs were identified in this study. Subsequently, we constructed a PPI network encoded by these DEGs and identified 10 genes closely associated with RA – LCK, GZMA, GZMB, CD2, LAG3, IL-15, TNFRSF4, CD247, CCR5, and CCR7. Furthermore, in the miRNA–hub gene networks, we screened out has-miR-146a-5p, which is the miRNA controlling the largest number of hub genes. Finally, we found some transcription factors that closely interact with hub genes, such as FOXC1, GATA2, YY1, RUNX2, SREBF1, CEBPB, and NFIC.
Conclusions
In summary, bioinformatics analyses were used to identify DEGs to find potential biomarkers that may be associated with RA. This study successfully predicted that LCK, FOXC1 and has-mir-146a-5p can be used as potential biomarkers of RA. Our study may have potential implications for future prediction of disease progression in symptomatic RA patients, and has important significance for the pathogenesis and targeted therapy of RA.