Abstract
Glioblastoma (GBM) is a malignant tumor of the central nervous system with high mortality rates. Gene expression profiling may determine the chemosensitivity of GBMs. However, the molecular mechanisms underlying GBM remain to be determined. To screen the novel key genes in its occurrence and development, two glioma databases, GSE122498 and GSE104291, were analyzed in the present study. Bioinformatics analyses were performed using the Database for Annotation, Visualization and Integrated Discovery, the Search Tool for the Retrieval of Interacting Genes, Cytoscape, cBioPortal, and Gene Expression Profiling Interactive Analysis softwares. Patients with recurrent GBM showed worse overall survival rate. Overall, 341 differentially expressed genes (DEGs) were authenticated based on two microarray datasets, which were primarily enriched in 'cell division', 'mitotic nuclear division', 'DNA replication', 'nucleoplasm', 'cytosol, nucleus', 'protein binding', 'ATP binding', 'protein C-terminus binding', 'the cell cycle', 'DNA replication', 'oocyte meiosis' and 'valine'. The protein-protein interaction network was composed of 1,799 edges and 237 nodes. Its significant module had 10 hub genes, and CDK1, BUB1B, NDC80, NCAPG, BUB1, CCNB1, TOP2A, DLGAP5, ASPM and MELK were significantly associated with carcinogenesis and the development of GBM. The present study indicated that the DEGs and hub genes, identified based on bioinformatics analyses, had significant diagnostic value for patients with GBM.