AI Parent-to-Metabolite Pathway Predictor
June 2026, ASMS Conference
Savannah M Mason1; Paula Cifuentes1, 2, 3; Tommaso Palomba1, 4; Ismael Zamora1
1Mass Analytica, S.L., Sant Cugat del Vallés, Spain; 2Universitat Pompeu Fabra, Barcelona, Spain; 3Lead Molecular Design, SL, Sant Cugat del Vallès, Spain; 4Molecular Discovery, Borehamwood, United Kingdom
Abstract
Introduction
Most drugs undergo chemical transformations in the body, known as biotransformations, to produce metabolites that are more readily eliminated. These reactions are largely mediated by metabolic enzymes, primarily in the liver, and exhibit high specificity, with each enzyme favoring particular substrates. Understanding the enzymes responsible for metabolite formation is critical for elucidating the metabolic pathways, predicting metabolic behavior, and anticipating potential toxicity. Metabolite Identification (MetID) studies, performed in vitro or in vivo, rely heavily on LC-MS/MS for the detection and structural identification of metabolites. However, most discovery studies provide limited information about the enzymes involved. Consequently, experimental approaches to reaction phenotyping, including recombinant enzymes incubations or chemical inhibition, are time- and resource-intensive, making comprehensive pathway characterization challenging.
Methods
This workflow integrates LC-MS MetID experiments from in vitro incubations. Users may apply a model to an experimentally identified metabolite to predict the possible enzymatic pathways responsible for its formation, including Phase I and Phase II reactions. The computational algorithm evaluates the exposure of reactive atoms of xenobiotic compounds to catalytic residues of human metabolic enzymes by simulating interactions between the two, using the enzyme’s 3D structure. Multiple docking poses are generated and scored based on energy contributions. The best pose is normalized to rank the probability, which is provided in the output. MetID experiments were analyzed using MassMetaSite in the ONIRO server with LC-MS data from Sciex and Thermo instruments. The predictions were performed using MetaSite 7 inside Oniro.
You must be logged in to access this content. Not yet registered? Create a new account