a pytorch implementation of Pafnucy.
Please find more about Pafnucy here.
Pafnucy is a 3D convolutional neural network that predicts binding affinity for protein-ligand complexes. It was trained on the PDBbind database and tested on the CASF "scoring power" benchmark.
The manuscript describing Pafnucy was published in Bioinformatics DOI: 10.1093/bioinformatics/bty374.
- Python 3.6+
- openbabel 2.4 [openbabel 3 has changed its API]
- numpy
- h5py
- pytorch
- matplotlib
- scikit-learn
- scipy
python main.py -i path/to/data
Data is provided in the original repo.