Identification of Key Genes via Integrated Multi-Omics and Machine Learning Uncovers Tumor Biological Features and Prognostic Biomarkers in Uterine Leiomyosarcoma - PubMed
14 hours ago
- #Tumor Immune Microenvironment
- #Machine Learning
- #Uterine Leiomyosarcoma
- Uterine leiomyosarcoma (ULMS) is a rare, aggressive uterine malignancy with high misdiagnosis rates and poor prognosis.
- Multi-cohort data integration (4 GEO datasets, TCGA-SARC, single-cell sequencing) identified 96 InteGenes enriched in cell cycle/p53/DNA repair pathways.
- 113 machine learning algorithms were compared, with the GBM model showing high diagnostic accuracy (training AUC = 1, validation accuracy 92.3-100%).
- 36 Mgenes (e.g., TRIP13, AUC = 0.972) demonstrated diagnostic value and correlation with the tumor immune microenvironment (TIME).
- Mgenes were found to modulate ULMS's TIME, with upregulated Mgenes linked to M2 TAMs/Tregs and downregulated Mgenes linked to effector cells.
- Mendelian randomization found no genetic causality between Mgenes and ULMS.
- The study advances understanding of ULMS molecular and immune features, offering potential diagnostic tools and immunotherapeutic targets.