The global proliferation of chemical compounds is linked with the release of pollutants into environmental waters, posing risks to ecosystems and human health. Managing this global issue requires a detailed understanding of the long-term availability and fate of pollutants and what impact they have on aquatic life. The safety of chemicals has traditionally been assessed via animal testing. However, the high costs and ethical concerns associated with these methods create a need for new methods to predict the risks associated with un-tested chemical compounds.
Previous work in the Grayson Group led to the development of a density functional theory (DFT)-based computational method for assessing the mutagenic risk of pharmaceutically important organic electrophiles (J. Chem. Inf. Model. 2019, 59, 5099). However, DFT calculations are time-consuming and expert-technical knowledge is required to perform them. In my project, machine learning (ML) models will be developed that can rapidly and easily predict reactivity descriptors for use in the prediction of aquatic toxicity. Collaboration with Dr Bryant, as a co-supervisor for the project, will focus on the mass-transport mechanisms by which organic compounds enter aquatic environments on catchment and system scales. This work will lead to a new ML protocol for computationally assessing the toxicity of pollutants and understanding what impact they will have on aquatic life.
In silico toxicology, Environmental chemistry, Machine learning, Chemical pollution
MRes Green Chemistry, Energy and the Environment – Imperial College London (Outstanding Performance Prize for highest marks in master’s cohort)
BSc Chemistry – University College London (Dean’s List Award for academic excellence)
Dr Matthew Grayson
Dr Pranav Singh
Dr Lee Bryant
