Identification of Disulfidptosis-Related Genes in Glioblastoma Combining Bioinformatics Techniques and Machine Learning

Abstract

Glioblastoma (GBM) is the most malignant brain tumor,  and with high heterogeneity and subsequent temozolomide (TMZ) resistance, the traditional treatment are often suboptimal. Disulfidptosis, a newly proposed form of cell death, plays a crucial role in the progression of various tumors, and its involvement in glioblastoma (GBM) remains further exploration. In this study, we combined bioinformatics with multiple machine learning algorithms to identify diagnostic and prognostic biomarkers associated with disulfidptosis in glioblastoma (GBM). First, we collected datas from TCGA and GTEx databases, and performed differential analysis, univariate Cox regression, and survival analysis. Additionally, we applied four machine learning algorithms to identify six candidate genes. We further investigated the potential biological functions and signaling pathways associated with these genes through GSEA and GSVA analysis. Subsequently, we constructed a ceRNA network and a drug-target network to explore their potential complex regulatory mechanisms. Finally, the expression levels of these genes were validated using clinical samples. Results show that these six candidate genes identified memorably, enriching in the JAK-STAT3 signaling pathway, were upregulated in glioblastoma (GBM) and were closely associated with patient prognosis. These findings may contribute to developing the molecular mechanisms underlying glioblastoma (GBM) and offer valuable implications for clinical diagnosis and treatment.