Chemical Reaction Systems
We look at chemistry as a collection of compounds behaving as systems. Rather than looking at chemistry molecule-by-molecule, we take a top-down approach, and investigate their collective behaviour. This view is heavily inspired by the way biological systems work. Experimentally speaking, flow chemistry and chromatographic reaction characterisation are key techniques in our lab’, and we call on methods from data science and machine learning to understand our data.
Machine Learning and Chemistry
Chemical properties and reactivity are complex, especially in mixtures of compounds. We are exploring the relationship between molecular structure, solution composition, and properties using machine learning and AI. We’re interested in how these models can be trained efficiently with experimental data, how they can provide useful predictions, and how they may be used to inform experimental investigations. Current projects involve the use of Chemical Language Models and graph-neural networks for the prediction of aqueous solubility and critical micellar concentration.