Identifying candidate genes involved in osteoarthritis through bioinformatics analysis.

This study aims to identify candidate genes and critical pathways involved in osteoarthritis (OA).Gene expression data of synovial membrane from OA patients and normal controls (NCs) were downloaded from database. Totally, 15 OA and 14 NC chips were available. Differentially expressed genes (DEGs) were identified through limma package (log2 fold change >0.585, false discovery rate (FDR) < 0.05), and protein-protein interaction (PPI) network was constructed using STRING. Moreover, perturbation and pathway enrichment analyses were performed through PerturbationAnalyzer in Cytoscape (iterative criteria <1×e-10) and clusterProfiler package (FDR <0.05), respectively.Totally, 236 up-regulated and 290 down-regulated DEGs were identified. In PPI network, 10 hub genes were found, including VEGFA, IL6, JUN, IL1B, ICAM1, ATF3, IL8, EGR1, CDKN1A, and JUNB. After perturbation analysis, 32 DEGs were passively and significantly changed, like PISD, RARRES3, EIF4G1, and EPHA3. Furthermore, 526 DEGs were enriched in 176 pathways, and pathway cross-talk network was constructed, involving 12 pathways and 66 cross-talks.Pathways like rheumatoid arthritis, osteoclast differentiation, and cytokine-cytokine receptor interaction might play critical roles in OA, and previously unreported genes VEGFA, JUN, JUNB, PISD, RARRES3, EIF4G1, and EPHA3 might participate in OA, providing novel directions for drug targeting.