A new generation of medicines, diagnostics and materials are needed to fuel new advances in healthcare and technology. Critical to this are predictive quality software tools to guide molecular design. In the School of Pharmacy at the University of Manchester, our constant aim is to improve the accuracy of the software tools we use in drug design. The aim of this proposal is to develop a more efficient and accurate software tool for prediction of protein-ligand binding affinities. This tool will use a simple but powerful new evolutionary computing method based on the cooperative behaviour of swarming insects. We have recently shown that this swarm-based approach dramatically improves the conformational optimisation of biomolecules. Now we propose to develop a tool based on a swarm of protein-ligand molecular dynamics (MD) trajectories which provides greater coverage of the high energy conformations, in turn leading to improved estimates of binding free energy. After validation, the tool will be used for predictive lead optimisation to impact our live oncology projects within the Manchester Cancer Research Centre (MCRC) and also in the industrial setting of AstraZeneca. The opportunity to confirm predicted protein inhibitors via synthesis and assay will also be provided.