Abstract
Gliomas, which are complex primary malignant brain tumors known for their heterogeneous and invasive nature, present substantial challenges for both treatment and prognosis. Recent advancements in whole-genome studies have opened new avenues for investigating glioma mechanisms and therapies. Through single-cell analysis, we identified a specific cluster of cancer cell-related genes within gliomas. By leveraging diverse datasets and employing non-negative matrix factorization (NMF), we developed a glioma subtyping method grounded in this identified gene set. Our exploration delved into the clinical implications and underlying regulatory frameworks of the newly defined subtype classification, revealing its intimate ties to glioma malignancy and prognostic outcomes. Comparative assessments between the identified subtypes revealed differences in clinical features, immune modulation, and the tumor microenvironment (TME). Using tools such as the limma R package, weighted gene co-expression network analysis (WGCNA), machine learning methodologies, survival analyses, and protein-protein interaction (PPI) networks, we identified key driver genes influencing subtype differentiation while quantifying associated outcomes. This study not only sheds light on the biological mechanisms within gliomas but also paves the way for precise molecular targeted therapies within this intricate disease landscape.