Jacques Gay

Development of new drugs involves billions of dollars, over many years, and yields approximately 20 new drugs a year with only 35% making it to clinical testing.^1,2 Our goal is to accelerate the early stages of drug discovery by improving the accuracy of computational binding free energy predictions. We will use a method combining machine learning and alchemical free energy perturbation (FEP) techniques called multimap targeted FEP³ to enable the calculation of binding free energies at the QM/MM accuracy. The methodology uses multiple invertible transformations to increase the overlap between states, thus accelerating the convergence of the estimate.

1. _{Mohs. R. C.; Greig. N. H. Drug discovery and development: Role of basic biological research. Alzheimer’s & dementia. 2017. 3(4), 651-657. DOI: https://doi.org/10.1016%2Fj.trci.2017.10.005} 
2. _{Borhani, D. W. The future of molecular dynamics simulations in drug discovery. J Comput Aided Mol Des. 2012, 26, 15-26. DOI: https://doi.org/10.1007/s10822-011-9517-y} 
3. _{Rizzi, A; Carloni, P; Parrinello, M.; PNAS, Free energies at QM accuracy from force fields via multimap targeted estimation. 2023. 120(46). DOI: https://doi.org/10.1073/pnas.2304308120}