Identification of key candidate genes and pathways in azoospermia by integrated GEO microarray database and bioinformatical analysis

Abstract

Objective To explore the pathogenesis of non-obstructive azoospermia at the molecular level, aiming to provide novel ideas for clinical diagnosis and treatment. Methods The R language was utilized to normalize the gene chip data of non-obstructive azoospermia downloaded from the Gene Expression Omnibus (GEO) and screen the differentially expressed genes. DAVID and KEGG databases were employed to carry out ontology function of the differentially expressed genes and enrichment analysis of the signaling pathways. A differentially-expressed gene co-expression network was delineated by using Cytoscape. The hub genes were calculated and identified by using CytoHubba. Enrichment analysis of the hub genes was performed by using ClueGo and Centiscape. Results A total of 518 differentially expressed genes were screened by R language, of which 271 genes were up-regulated and 247 were down-regulated. The ontology function analysis and enrichment analysis of signaling pathways prompted that these differentially expressed genes played a pivotal role in the biological processes of spermatogenesis, sperm chromatin condensation, formation of sperm acrosome membrane and vesicle, sperm-egg recognition, cell differentiation, ATP coupling and transcription factor binding, etc. Analysis of the differentially expressed gene co-expression network demonstrated that the hub genes including GAPDHS, PCSK4, TSSK1B and TSSK2 played a key role in the spermatogenesis and differentiation processes. Conclusion Exploration of the GEO gene chip data from multiple dimensions and systematical analysis of the internal information are of significance to identify the molecular mechanism underlying non-obstructive azoospermia. Key words: Azoospermia; Database; Bioinformatics; Sperm; Gene; Enrichment analysis