In drug discovery, when the structure of a target protein is know ‘docking’ generates many ligand conformations/poses and prioritizes them on their binding-affinity to the target.
Docking suffers from under-sampling and the use of approximate scoring functions, and therefore misses many promising ligands.
Docking quality can be improved by generating more ligand-target configurations or by increasing the quality of the scoring function, both directions are limited by the available computational resources.
Finding the optimal input (configuration) of the underlying structure-affinity relationship (SAFIR) is difficult because SAFIR is a hidden function.
PASQAL has developed a unique and proprietary quantum powered optimization method which efficiently finds the input (configuration) which extremizes the output (binding affinity) of a hidden function (SAFIR) where that input can be both discrete as well as continuous and we tested this method on a toy model molecular structure optimization.