Background: This study aimed to identify glioblastoma prognosis-associated genes with potential diagnosis or prognosis values using integrated bioinformatics analysis. Results: In total, 1831 differentially expressed genes (DEGs) between the glioblastoma and control samples were identified and were clustered into seven weighed gene co-expression network analysis (WGCNA) modules. These DEGs were associated with different functional categories and pathways. Nine prognosis-associated DEGs (including glutaminase 2 [GLS2] and neutrophil cytosolic factor 2 [NCF2]) were identified, and the higher expression levels of GLS2 and NCF2 genes were associated with the poor prognosis of glioblastoma in 'The Cancer Genome Atlas' cohort and a clinical cohort. Conclusion: These results showed that the two genes play novel roles in the etiological and development of glioblastoma.
Keywords: Cox regression analysis; differentially expressed genes; glioblastoma; prognosis-associated genes; survival analysis.