Integrative machine learning-driven prognosis and immunotherapy stratification via lactylation-associated gene in ovarian cancer.

ABSTRACT: This study utilized a comprehensive approach by integrating multi-omics data to systematically assess lactate modification levels across diverse cell types employing AUCell, JASMINE, and singscore algorithms. An epithelial subpopulation exhibiting the highest lactylation score was successfully pinpointed, and differentially expressed genes linked to lactylation were identified. Through machine learning techniques, a prognostic model was developed based on three genes (TMEM126B, PYGL, and NDUFS6). This model displayed significant associations with immune tumor microenvironment characteristics, microsatellite instability, immune checkpoint expression, and tumor mutation burden. Elevated lactylation risk was linked to the activation of cell cycle and oncogenic pathways, dampened anti-tumor im

SUBMITTER: Li X 

PROVIDER: S-EPMC12635173 | biostudies-literature | 2025 Nov

REPOSITORIES: biostudies-literature