Provides an approach for the generation of ligand images filling protein pockets based on deep neural networks. LigVoxel produces predictions that are responsive to the number of atoms selected as input. These predictions significantly overlap with ligand features of previously unseen ligands, and they can be used to select poses and conformers close to the native ligand orientation and geometry.
Computational Biophysics Laboratory, Universitat Pompeu Fabra, Barcelona Biomedical Research Park, Barcelona, Spain; Acellera, Barcelona Biomedical Research Park (PRBB), Barcelona, Spain; Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain
LigVoxel funding source(s)
Supported by Acellera Ltd., MINECO (BIO2017-82628-P), FEDER, and the European Union’s Horizon 2020 research and innovation programme under grant agreement No 675451 (CompBioMed project).