Abstract
BACKGROUND: Increasing attention has been drawn to the close association between exposure to environmental endocrine-disrupting chemicals (EDCs) and the risk of acute myocardial infarction (AMI). This study aimed to utilize multi-omics approaches to identify EDCs-related diagnostic candidate and elucidate their potential roles in AMI. METHODS: This study integrated toxicological targets of 13 representative EDCs with transcriptomic datasets from AMI cohorts. Machine learning algorithms (LASSO and SVM-RFE) were employed to screen feature genes, and model interpretability was achieved via SHAP analysis. Additionally, Summary-data-based Mendelian Randomization (SMR) and clinical validation in plasma samples were utilized to verify the candidate. RESULTS: A total of 1,818 potential toxicological targets for EDCs were screened. The machine learning algorithms identified an 8-gene diagnostic signature, which demonstrated superior predictive performance in an external validation set (AUC=0.968). Through intersection analysis, MERTK was pinpointed as a critical EDCs-related mediator and exhibited a strong positive correlation with the infiltration of immune cells, such as monocytes and neutrophils. SMR results further confirmed that MERTK is a causal risk driver for AMI. Consistently, clinical validation demonstrated that plasma MERTK levels were significantly elevated in AMI patients compared to controls. CONCLUSION: This study established a multi-omics model linking EDCs exposure to AMI pathogenesis and identified MERTK as a key diagnostic candidate potentially mediating EDC-induced AMI. These findings provide novel molecular targets for environmental risk assessment and precision diagnosis.