Objective: This study bioinformatically analyzed aberrant genes and pathways for associations with glioblastoma development and prognosis.
Methods: The Gene Expression Omnibus (GEO) database was searched and 4 GEO datasets (GSE4290, GSE50161, GSE116520, and GSE90598) were retrieved for limma and RobustRankAggreg package analyses of differentially expressed genes (DEGs) between glioblastoma and normal brain tissues. Functional enrichment analysis was conducted for the main biological functions of these DEGs, whereas the hub genes were identified using the protein-protein interaction network and confirmed for transcriptional and translational levels using the Cancer Genome Atlas, the Genotype-Tissue Expression, and the Human Protein Atlas data. The prognostic values of these hub genes were analyzed using the Chinese Glioma Genome Atlas. Their transcriptional factor regulation network was constructed to assess the roles in glioblastoma development and progression.
Results: A total of 473 DEGs (182 upregulated and 291 downregulated) were identified and the hub genes (including CCNB1, CDC20, CCNB2, BUB1, and CCNA2) were shown in module 1 and enriched in the cell cycle or p53 signaling pathway. The highly expressed CCNB1, CDC20, BUB1, and CCNA2 in patients with glioblastoma were associated with poor overall survival, whereas TAF7 could upregulate expression of CCNB1 and CCNA2 and GTF2E2 could upregulate CDC20 expression in glioblastoma.
Conclusions: This study showed several DEGs in glioblastoma, and aberrant expression of their hub genes was associated with glioblastoma pathogenesis and poor prognosis, especially the signaling axes of TAF7/CCNB1, TAF7/CCNA2, and GTF2E2/CDC20.
Keywords: Bioinformatical analysis; CGGA; Differentially expressed genes; GEO; Glioblastoma; TCGA; Transcriptional factors.
Copyright © 2020 Elsevier Inc. All rights reserved.