Integrated network toxicology, machine learning, and bioinformatics analysis reveals sodium dehydroacetate-induced coagulation dysfunction in colorectal cancer

作者：Zhenzhen Yang, Cheng Chang, Liya Jiang, Ying Liu, Na Gao, Xiaolin Zhang, Yinsen Song, Tianli Fan · 发表于：Drug and Chemical Toxicology · 年份：2026 · DOI：10.1080/01480545.2025.2606908 · 被引用次数：1 · 研究领域：Cancer, Hypoxia, and Metabolism、Alcohol Consumption and Health Effects、Metabolomics and Mass Spectrometry Studies

Coagulation dysfunction, a common hematologic disorder with unclear pathogenesis, is influenced by environmental factors. Sodium dehydroacetate (SDA), a widely used preservative with high environmental mobility and persistence, has become an emerging organic contaminant and is increasingly recognized for its potential to disrupt immune homeostasis and induce coagulation abnormalities, yet its specific mechanisms remain poorly understood. In this study, we employed an integrated computational approach-combining network toxicology, machine learning (LASSO and XGBoost), bioinformatics, molecular docking, and molecular dynamics simulations-to systematically investigate SDA-induced coagulation dysfunction. We identified 191 potential targets, with significant enrichment in cancer-related pathways, atherosclerosis, and proteoglycans in cancer. Met proto-oncogene (Met) emerged as a core target through machine learning. Analysis of a colorectal cancer dataset (GSE52060) revealed elevated Met expression in patients with coagulation dysfunction, and receiver operating characteristic analysis indicated its strong diagnostic value (area under the curve = 0.856). Molecular docking showed stable binding between SDA and Met (-5.5 kcal/mol), further supported by molecular dynamics simulations demonstrating favorable hydrogen bonding and complex stability. This study provides a theoretical foundation for understanding SDA's role in coagulation dysfunction and supports future preventive and th...