Identification of ammonia death-related prognostic genes in hepatocellular carcinoma through integrated machine learning and transcriptomic analysis
10.3969/j.issn.1009-9905.2025.07.007
- VernacularTitle:结合机器学习和转录组学分析筛选与氨死亡相关的肝癌预后基因
- Author:
Li-yan JIA
1
;
Bai-hong ZHENG
;
Guo-hao WANG
;
Xiu-wen GUO
;
Ying WANG
Author Information
1. 山东大学第二医院第二手术部(山东 济南 250033)
- Publication Type:Journal Article
- Keywords:
Ammonia-induced cell death;
Hepatocellular carcinoma;
Machine learning;
Prognostic biomarker
- From:
Chinese Journal of Current Advances in General Surgery
2025;28(7):545-551
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To systematically evaluate the molecular characteristics and prognostic value of ammonia death-related genes in hepatocellular carcinoma(HCC).Methods:Consensus unsupervised clustering was used to identify ammonia death-related molecular subtypes in HCC samples.Weighted gene co-expression network analysis(WGCNA)was applied to identify gene modules associated with ammonia death.Support vector machine(SVM)and LASSO algorithms were used to screen four hub genes,and a risk score system was constructed based on a LASSO-Cox regression model.The association between the risk model and patient survival,tumor microenvironment,and re-sponse to immunotherapy was further analyzed.Results:Consensus clustering identified two distinct ammonia death-related molecular subtypes(P<0.05).The constructed risk score model showed good predictive performance for overall survival in HCC patients and was closely associated with immune infiltration characteristics of the tumor microenviron-ment and immunotherapy responsiveness(P<0.05).Conclusion:The ammonia death-related risk score model may serve as a novel prognostic biomarker for HCC and provide potential guidance for immunotherapy strategies.
