Multiple bioinformatics analyses of integrated gene expression profiling data and verification of hub genes associated with diabetic retinopathy

Abstract

Background: Diabetic retinopathy (DR) is a serious complication of diabetes that can lead to blindness. This study aimed to identify the core genes and molecular functions involved in DR through multiple bioinformatics analyses. Material/Methods: The mRNA gene profiles of human DR tissues from the GSE60436 and GSE53257 datasets were assessed with R software and integrated to identify the co-expressed differentially expressed genes (DEGs). Multiple bioinformatics analyses were used: Gene Ontology (GO) analysis, signaling pathway analysis, and hub gene prediction. Quantitative reverse transcription-PCR (qRT-PCR) was used to verify the hub genes. Results: The Database for Annotation, Visualization and Integrated Discovery (DAVID) online tool suggested that the biological processes of the DEGs focused on mitochondrial transport, the cellular components focused on mitochondria, and molecular functions focused on catalytic activity. The results provided by DAVID were consistent with those provided by STRING and the GeneMANIA online database. All the DEGs function in metabolic pathways, consistent with the g: Profiler online analysis results. The protein-protein interaction (PPI) networks forecasted by STRING and GeneMANIA were entered into Cytoscape for cytoHubba degree analysis. The hub genes predicted by cytoHubba suggested that fumarate hydratase (FH) might be relevant to DR. qRT-PCR suggested that the expression of FH was higher in DR retinal tissues than in normal control tissues. Conclusions: Multiple bioinformatics analyses verified that FH could be used as a potential diagnostic marker and new therapeutic target of DR.