Transformation-Aware Molecular Networking for Interpretation of Untargeted LC–HRMS Data

The rapid emergence of new psychoactive substances (NPS) and their extensive biotransformation challenge the reliability and interpretability of chemical measurements in forensic toxicology. Targeted analytical workflows offer high selectivity but frequently fail when parent compounds are present at low concentrations or are absent from biological matrices, limiting metabolite coverage and evidential interpretation. In this study, an untargeted LC–HRMS measurement and data-analysis framework was evaluated to improve metabolite annotation and structural contextualization in complex forensic samples. Blood and urine collected from a suspected driving under the influence of drugs (DUID) case were analyzed using high-resolution full-scan and data-dependent MS/MS acquisition as a representative test system. Data were processed in Compound Discoverer using a customized workflow combining spectral library matching (mzCloud and in-house spectral database built from HighResNPS.com, containing over 2400 high-resolution spectra of drugs of abuse and NPS), rule-based metabolite prediction (MetID), and transformation-aware molecular networking. The transformation-aware molecular networking strategy integrates MS/MS spectral similarity with predicted phase I and phase II biotransformations, enabling relational organization of parent compounds, metabolites, and structurally related features. Compared to rule-based metabolite prediction alone, this approach increased the number of metabolite-related features associated with detected xenobiotics and supported reconstruction of chemically consistent metabolic families, including cases in which parent compounds were not observed in the measured sample. Taken together, the results show that transformation-aware molecular networking provides an effective means of organizing and interpreting untargeted LC–HRMS data by linking spectral similarity with biotransformation relationships. This framework supports a more contextual interpretation of metabolite-related features in forensic and toxicological investigations involving NPS and other xenobiotics.