Metabolic profiling of Yangxinshi tablet based on time-staggered ion list dynamic detection integrated with metabolic molecular network.

A comprehensive analytical strategy was developed to systematically characterize the in vivo metabolic profiles of chemical compounds in Yangxinshi Tablet (YXST) after oral administration in rats. First, an innovative data acquisition mode, termed background subtraction-assisted virtual polygonal mass defect filter with time-staggered preferred ion list and active exclusion (BS-VPMDF-tsPIL-AE), was established for biological sample data acquisition. Second, the metabolic molecular network (MMN) was used to automatically map metabolic pathways and prototype-metabolite relationships, enabling rapid annotation of biotransformation derivatives. Third, deep learning-assisted mass defect filtering (MDF) intelligent classification effectively supported metabolite types identification. Finally, a total of 134 drug-derived components (75 prototypes and 59 metabolites) and 165 drug-derived components (83 prototypes and 82 metabolites) were identified in rat plasma and urine, respectively. Additionally, 29 and 31 prototype components were preliminarily identified in plasma and urine by comparison with standards. By amalgamating virtual polygonal mass defect filter (VP-MDF) and time-staggered preferred ion list (tsPIL), the presented approach effectively enhances MS/MS coverage and minimizes interference from co-eluted metabolites, thereby addressing the challenge of detecting low-abundance drug-derived components in complex biological matrices. Combining R programming, Python, and MMN-driven visualization, it provides a powerful platform for comprehensive acquisition and interpretation of in vivo metabolic fingerprints of natural products, advancing research on the material basis of traditional Chinese medicine (TCM) efficacy.