8 datasets found

A dataset of 3D structures of about 10k complexes, i.e., structures of ligands docked in binding pockets of proteins.

A dataset of 102 proteins, 22,886 active compounds, and over 1M decoys for training and testing.

A new benchmark dataset for structure-based virtual screening, constructed based on DUD-E and PDBBind databases.

CrossDocked 2022 dataset for structure-based virtual screening, used for benchmarking docking methods.

The MOSES dataset is a large dataset of molecules, and the HIV dataset is a subset of the MOSES dataset containing 41127 molecules.

The LIT-PCBA dataset contains experimentally confirmed active and inactive compounds for 15 targets.

Virtual screening, which identifies potential drugs from vast compound databases to bind with a particular protein pocket, is a critical step in AI-assisted drug discovery....

The dataset used in the paper is the Ligand-Based Virtual Screening dataset, which is a large-scale dataset for virtual screening.