Analysis of differential gene expression caused by cervical intraepithelial neoplasia based on GEO database.

Abstract

The aim of the present study was to identify the differentially expressed genes between cervical intraepithelial neoplasias (CIN) and adjacent normal tissue, and to construct a protein-protein interaction (PPI) network. A CIN dataset was obtained from Gene Expression Omnibus, and data of gene expression in CIN and adjacent normal tissue were extracted from GSE64217. The differentially expressed genes were selected using software package and heat map was drawn using the ‘pheatmap’ package. The selected differentially expressed genes were subjected to PPI, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis using Cytoscape, Database for Annotation, Visualization and Integrated Discovery, STRING and KOBAS. In the present study, 287 genes were differentially expressed between CIN and adjacent normal tissue, of which 170 were significantly upregulated and 118 genes were significantly downregulated (P<0.00001, fold-change >6). A differential gene expression network map was constructed to show the interactions of 30 protein products encoded by differentially expressed genes using STRING software. In particular, the key gene, EGR1, was identified using Cytoscape software. The KEGG pathway analysis revealed that the differential genes were mainly involved in several pathways, including ‘glutathione metabolism’, ‘arachidonic acid metabolism’, and ‘pentose phosphate pathway’. Results of the GO analysis showed that differential genes were enriched in different subsets. Specifically, small proline-rich protein 2E and 3, distal-less homeobox 5, epithelial membrane protein 1, cornifelin, periplakin, homeobox protein Hox-A13, estrogen receptor α, transglutaminase 1, small proline-rich protein 2A, Rh C glycoprotein, tumor protein p63, TGM3, homeobox B5 and small proline-rich protein 2D were enriched in ‘epithelial cell differentiation’, which affected the differentiation of epithelial cells. In conclusion, 287 differentially expressed genes were identified successfully. The key gene was identified based on the results of PPI, GO and KEGG analyses, and functional annotation and pathway analysis were also performed. Our study provides the basis for further studies on the interaction among differentially expressed genes.