Fragment-based drug discovery (FBDD) is an efficient methodology for the generation of lead compounds. However, structure-based optimisation of weakly binding fragments into high affinity drug molecules often means that difficult synthetic challenges arise, hence there is a growing demand for cutting-edge synthetic methods and experimental techniques. This project focusses on machine-learning methodologies to more effectively plan and execute synthetic sequences to obtain desired target molecules mandated by FBDD optimisation.
Taylor et al., “Accelerated Chemical Reaction Optimization using Multi-Task Learning”; ACS Cent. Sci. 2023 https://doi.org/10.1021/acscentsci.3c00050
Zakrzewski et al., “Scalable Palladium-Catalyzed C(sp3)–H Carbonylation of Alkylamines in Batch and Continuous Flow”; Org. Process Res. Dev. 2023 https://doi.org/10.1021/acs.oprd.2c00378
Taylor et al., “A Brief Introduction to Chemical Reaction Optimization”; Chem Rev. 2023 https://doi.org/10.1021/acs.chemrev.2c00798
Pomberger et al., “The effect of chemical representation on active machine learning towards closed-loop optimization”; React. Chem. Eng. 2022 https://doi.org/10.1039/D2RE00008C