Clinica chimica acta; international journal of clinical chemistry · 2025
Lipidomic machine learning predictor for progression of gastric cancer.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Ding Xian, Zhou Jun, Wang Zhaoying, Bo Yunhai, Abliz Zeper, An Zhuoling, Yang Fen, Chen Yanhua
Аннотация
BACKGROUND: Gastric cancer (GC) is a leading cause of global cancer mortality, characterized by aggressive progression and poor prognosis. The development of robust strategies to identify and characterize GC disease progression is urgently needed to improve patient outcomes.
METHODS: We employed an integrated analytical framework, combining machine-learning-driven lipidomics and Mendelian randomization (MR), to identify and validate lipid biomarkers associated with GC tumorigenesis and metastatic progression.
RESULTS: Our analysis revealed profound lipidomic dysregulation, with 57.9 % of detected lipids significantly altered during disease progression. Different lipid species exhibited distinct dynamic patterns: sphingomyelins (SMs), fatty acids (FAs), and ceramides (Cers) accumulated progressively, while triacylglycerols (TGs), phosphatidylcholines (PCs), phosphatidylethanolamines (PEs), and lysophosphatidylcholines (LPCs) were consistently depleted. Network analysis highlighted critical roles for phospholipid remodeling in GC pathogenesis, involving PI, PE, PC, and SM species. MR and clinical correlation analysis established causal relationships for 11 lipid species with GC risk. Leveraging profiles of 11 lipids, we developed a high-performance machine-learning predictor, achieving 92.3 % overall accuracy for GC detection and 78.3 % accuracy for metastasis prediction.
CONCLUSIONS: This study provides fundamental insights into lipid metabolic reprogramming during GC progression and establishes a clinically actionable predictive platform. Our findings offer a valuable resource for guiding future research and therapeutic strategies targeting GC metastasis.
METHODS: We employed an integrated analytical framework, combining machine-learning-driven lipidomics and Mendelian randomization (MR), to identify and validate lipid biomarkers associated with GC tumorigenesis and metastatic progression.
RESULTS: Our analysis revealed profound lipidomic dysregulation, with 57.9 % of detected lipids significantly altered during disease progression. Different lipid species exhibited distinct dynamic patterns: sphingomyelins (SMs), fatty acids (FAs), and ceramides (Cers) accumulated progressively, while triacylglycerols (TGs), phosphatidylcholines (PCs), phosphatidylethanolamines (PEs), and lysophosphatidylcholines (LPCs) were consistently depleted. Network analysis highlighted critical roles for phospholipid remodeling in GC pathogenesis, involving PI, PE, PC, and SM species. MR and clinical correlation analysis established causal relationships for 11 lipid species with GC risk. Leveraging profiles of 11 lipids, we developed a high-performance machine-learning predictor, achieving 92.3 % overall accuracy for GC detection and 78.3 % accuracy for metastasis prediction.
CONCLUSIONS: This study provides fundamental insights into lipid metabolic reprogramming during GC progression and establishes a clinically actionable predictive platform. Our findings offer a valuable resource for guiding future research and therapeutic strategies targeting GC metastasis.