Abstract

Data-driven, Non-targeted Metabolomic Approaches Using High Resolution Mass Spectrometry and Advanced Machine Learning Algorithms

Manfred Beckmann, Research Fellow, University of Wales Aberystwyth

Non – targeted metabolomic analysis using High Resolution Flow Infusion Electrospray ionisation - Mass Spectrometry (FIE-HRMS) provides comprehensive and un-biased metabolite fingerprints of biological samples. The Aberystwyth High Resolution Metabolomics Laboratory (HRML) combines advanced machine learning algorithms, such as random forest, with high performance computing clusters and sensitive OMICS analytical instruments to elucidate metabolic changes in a range of biological systems; from profiling bio – energy grasses, host – pathogen interactions in model crops and habitual and acute food consumption biomarker discovery in human biofluids. Potentially discriminatory features are elucidated using Ultra High Performance Liquid Chromatography-High Resolution MS (UHPLC-HRMS), using both Reverse Phase C18 and Hydophillic Interaction Chromatography (HILIC), to cover a range of metabolite chemistry. The combination of chromatography, accurate mass and tandem mass spectrometry (MSn) with standards means that we can accurately investigate features after modelling, without the need for extensive targeted studies. This talk highlights the utilisation of our methods and workflows in the field of metabolomics and their application to aid in the understanding of a wide range of biological systems.