This study was aimed to identify hub genes associated with the development of glioblastoma (GBM) by conducting a bioinformatic analysis. The raw gene expression data were downloaded from the Gene Expression Omnibus database and The Cancer Genome Atlas project. After the differentially expressed genes (DEGs) were identified, the functional enrichment analysis of DEGs was conducted. Subsequently, the protein-protein interaction (PPI) network, molecular complex detection clusters, and transcriptional factor (TF)-miRNA-target regulatory network were constructed, respectively. Furthermore, the survival analysis of prognostic outcomes and genes was analyzed. In addition, the expression of key genes was validated by quantitative real-time PCR (qRT-PCR) analysis. A total of 884 DEGs, including 418 upregulated and downregulated genes, were identified between GBM and normal samples. The PPI network comprised a set of 3418 pairs involving 751 nodes, and AKT1 and CDK2 were the critical genes in the network. A total of seven clusters were identified, the genes in which were intensively associated with cell cycle, cholinergic synapse, and extracellular matrix (ECM)-receptor interaction. qRT-PCR analysis indicated that AKT1 and CDK2 were significantly upregulated, and NRXN3 and NPTX2 were significantly downregulated in GBM samples. The TF-miRNA-target regulatory networks were built, in which CCNB1, RFC5, microRNA524, and microRNA34b were key regulators. There were 43 genes, including NPTX2 and NRXN3, significantly related to the prognostic outcomes of GBM patients. These crucial genes might be promising options for GBM treatment.
Keywords: abnormal expression genes; glioblastoma; hub genes; prognosis; regulatory networks.