ChEMBL234 dataset, originally part of ChEMBL database [1], processed in MoleculeACE [2] for activity cliff evaluation. It is intended to be use through scikit-fingerprints library.
The task is to predict the inhibitor constant (Ki) of molecules against the D(3) dopamine receptor target.
Characteristic
Description
Tasks
1
Task type
regression
Total samples
3657
Recommended split
activity_cliff
Recommended metric
RMSE
References… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/MoleculeACE_chembl234_ki.