Examining LCK and BDNF as Potential Prognostic Biomarkers in Glioblastoma multiforme: Insights from The Cancer Genome Atlas (TCGA)Transcriptomic Analysis
Examining LCK and BDNF as Potential Prognostic Biomarkers in Glioblastoma multiforme: Insights from The Cancer Genome Atlas (TCGA)Transcriptomic Analysis
Saeid Latifi-Navid,1,*MohammadAli Shahmohammadi,2Fatemeh Hedayat,3Seyedeh Azin Azad Abkenar,4
1. Department of Biology, Faculty of Sciences, University of Mohaghegh Ardabili, Ardabil 56199-11367, Iran 2. Department of Biology, Faculty of Sciences, University of Mohaghegh Ardabili, Ardabil 56199-11367, Iran 3. Department of Biology, Faculty of Sciences, University of Mohaghegh Ardabili, Ardabil 56199-11367, Iran 4. Department of Biology, Faculty of Sciences, University of Mohaghegh Ardabili, Ardabil 56199-11367, Iran
Introduction: Introduction: Glioblastoma multiforme (GBM) is a rare but highly malignant brain tumor with a global incidence of fewer than 10 cases per 100,000 people [1]. Despite treatment advances, it remains largely incurable, with a prognosis of only 14-15 months survival post-diagnosis [2]. In this study, we sought to identify genes that impact GBM survival rates in order to uncover new prognostic biomarkers for improved prognosis prediction.
Methods: Methods: In this study, we utilized 'TCGAbiolinks' to access raw RNA-seq data (STAR-Count) and clinical information for Glioblastoma Multiforme (GBM) from TCGA. We identified differentially expressed genes (DEGs) using 'TCGAanalyze_DEA' in combination with edgeR functions, applying stringent criteria (|LogFC| ≥ 2 and FDR = 0.01) for DEG selection. DEGs were organized in a structured table using 'TCGAanalyze_LevelTab,' displaying expression levels in tumor (Cond1type) and normal (Cond2type) samples, along with the 'Delta' value. We enhanced interpretability by replacing gene Ensemble IDs with protein-coding gene names via the 'biomaRt' package and ensured data integrity by eliminating duplicate gene names. Enrichment analysis with 'TCGAanalyze_EAcomplete' revealed associations between DEGs and Gene Ontology (GO) functions and KEGG pathways, identifying over-represented genes or proteins among up-regulated genes. We constructed a protein-protein interaction (PPI) network with the STRING database, analyzing it with Cytoscape, focusing on degree centrality for node evaluation. Finally, we employed the UALCAN database to generate Kaplan-Meier survival curves for hub genes. This streamlined approach allowed us to comprehensively analyze GBM data, identifying significant genes and pathways.
Results: Results: In an analysis of 175 patients and 60,660 genes, 6,905 Differentially Expressed Genes (DEGs) were identified, with 4,386 up-regulated and 2,519 down-regulated in tumor samples compared to normal. These DEGs were categorized by GO functions into Molecular Functions (MF), Biological Processes (BP), and Cellular Components (CC). In MF, DEGs were significantly associated with channel activity (GO:0015267) with a false discovery rate (FDR) of 1.49e-31. In CC, a notable enrichment of DEGs was found in the plasma membrane part (GO:0044459, FDR 5.67e-62). In BP, cell-cell signaling (GO:0007267) was significantly enriched with an FDR of 6.32e-23. Additionally, KEGG analysis highlighted DEGs' enrichment in the cAMP-mediated signaling pathway (FDR 1.04e-11). Using cytoHubba, the top 10 degree nodes were identified, with only two showing a significant association with poor prognosis based on Kaplan-Meier and log-rank test analysis (P<0.05): LCK proto-oncogene (LCK) and brain-derived neurotrophic factor (BDNF).
Conclusion: Conclusions: This study has revealed driver genes and pathways in Glioblastoma multiforme, which may serve as biomarkers and therapeutic targets.