Constructing a Glioblastoma Prognostic Model Related to Fatty Acid Metabolism Using Machine Learning and Identifying F13A1 as a Potential Target.

ABSTRACT: Background: Increased fatty acid metabolism (FAM) is an important marker of tumor metabolism. However, the characterization and function of FAM-related genes in glioblastoma (GBM) have not been fully explored. Method: In the TCGA-GBM cohort, FAM-related genes were divided into three clusters (C1, C2, and C3), and the DEGs between the clusters and those in the normal group and GBM cohort were considered key genes. On the basis of 10 kinds of machine learning methods, we used 101 combinations of algorithms to construct prognostic models and obtain the best model. In addition, we also validated the model in the GSE43378, GSE83300, CGGA, and REMBRANDT datasets. We also conducted a multifaceted analysis of F13A1, which plays an important role in the best model. Results: C2,

SUBMITTER: Liu Y 

PROVIDER: S-EPMC11852379 | biostudies-literature | 2025 Jan

REPOSITORIES: biostudies-literature