Profiling the metabolites present in a given sample (e.g., blood, urine) enables researchers to investigate various internal (e.g., disease, microbiome) and external (e.g., exposome, drugs) perturbations. Developing innovative approaches to shed light on patients’ metabolic profile is of critical importance to understand the events triggering or modulating a particular trait of interest (e.g., disease progression, drug metabolism, chemical risk assessment). Currently, the majority of metabolomics studies do not consider the chemical structure of trait-associated metabolites. In this project, we discuss a novel cheminformatics-based approach to integrate, visualize, and model such type of complex metabolomics dataset. We considered a case study consisting of metabolite profiles for 52 adenocarcinoma patients and 31 healthy controls. By characterizing metabolite chemical structures using both computed molecular descriptors and measured metabolic profiles, we enhance the understanding of metabolite features associated with cancer and develop strategies for identifying novel biomarkers. Functional enrichment analysis to discover cancer-associated metabolites is conducted using groups of metabolites defined based on metabolites’ pairwise structural similarity. This new approach could help researchers in structurally mining, comparing, and analyzing metabolites by focusing on the metabolites’ structural features and chemical properties. We will show how this approach can facilitate the biological understanding and advance research based on metabolomics data.

Cheminformatics approach to exploring and modeling trait-associated metabolic profiles

254th National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.