James, is a research staff member at IBM Research UK, and part of the IBM research presence at the Science and Technology Facilities Council (STFC) in Daresbury UK.
James completed his doctoral studies at the University of St Andrews, Scotland UK, where his research focused on molecular property prediction, using a state of the art combinations of quantum chemistry and machine learning methodologies. This research concentrated on the prediction of solubility, which is a critical parameter for determining the biological effects of a compound.
Following the successful completion of his PhD, he moved to the University of Manchester, to undertake research related to the creation of a novel polarizable force field (FFLUX) for biomolecular simulation. FFLUX, combines quantum mechanical energy partitioning, Interacting Quantum Atoms (IQA), with machine learning, aiming to provide an efficient and accurate methodology. This work focused on expanding the existing IQA theory and methodology to enable the partitioning of post-Hartree-Fock wave functions, and determine the energy contributions of dynamic electron correlation to chemical bonding. Having completed this, James explored machine learning predictions of dynamic electron correlation energy contributions.
James' current work focuses on developing and applying molecular calculation methods, for modelling industrially relevant systems. This work combines work with molecular simulation methods, machine learning and high performance computing. His primary research interests relate to the modelling of condensed matter, molecular property prediction and machine learning.
More of James' current work can be found on the chemistry group website