LINGO1, C7orf31, and VEGFA are prognostic genes of primary glioblastoma: analysis on gene expression microarray
Abstract:
Glioblastoma is the most prevalent malignant glioma in WHO grade IV and its median overall survival is 12-15 months. This study identifies the primary glioblastoma. prognostic genes. Gene expression data in primary glioblastomas with short-term (<12 months, N=16), intermediate (12-36 months, N=31), and long-term (>36 months, N=23) overall survival were downloaded from Gene Expression Omnibus (GSE53733). Limma determined the differentially expressed genes (DEGs) between different groups (|log2 fold change| ≥0.5 and p-value <0.05). STRING database and Cytoscape were used to predict protein-protein interactions between DEGs and to construct the PPI network (PPI, medium confidence ≥0.4), and CytoNCA plugin in Cytoscape calculated each DEG’s degree, betweenness, sub-graph and closeness centralities. Long-term/short-term survival-related DEGs were defined as those with increased/decreased expression values and survival time. The following DEGs were identified; 161 between intermediate and short-term glioblastomas, 465 between long-term and short-term and 624 between long-term and intermediate tumors. The common FLRT1 and LINGO1 up-regulated DEGs and common down-regulated C7orf31 were identified in these three DEG sets. PPI networks were established, and VEGFA was the key DEG in each PPI network. The short-term survival-related DEGs were enriched in 3 cancer-related pathways. Moreover, FLRT1 and LINGO1 were long-term survival-related DEGs and C7orf31 and VEGFA were short-term survival DEGs. LINGO1, C7orf31, and VEGFA were confirmed using a further dataset, and we therefore conclude that LINGO1 might be a positive primary glioblastoma prognostic gene and C7orf31 and VEGFA might be negative prognosticators.
Received date: 07/22/2017
Accepted date: 11/03/2017
Ahead of print publish date: 07/30/2018
Issue: 4/2018
Volume: 65
Pages: 532 — 541
Keywords: Differentially expressed gene, overall survival, primary glioblastoma, prognosis, protein-protein interaction
DOI: 10.4149/neo_2018_170722N496