Creation of a vendor independent HRMS processing workflow for identifying emerging chemical threats

Current methods to detect chemicals of concern in non-targeted HRMS data is limited either because the scope of compounds searched for is limited, or by the amount of work required in searching a large number of compounds. This project proposes an algorithm which automizes feature detection, componentisation, searches for matches via a universal library, and processes unmatched or unlikely matches for class detection via a machine learning algorithm. Through this, HRMS repositories can be processed, and through their metadata the spatial and temporal characterisation of emerging chemical threats can be identified.

Mathieu Feraud completed his masters in Data Science in 2019 at The University of Queensland. His capstone project, which was in collaboration with QAEHS, developed a machine learning model to predict toxicity based on the liquid chromatography high-resolution mass spectrometry data and existing toxicity data. In 2020 he has been working on the ARC Discovery project to develop a global platform for identifying emerging chemical threats. Mathieu started his PhD in October 2020 in which he will study and develop predictive modeling.

Please note this is a PhD student confirmation milestone review.