compassdock 0.1.2

A comprehensive accurate assessment approach for deep learning based molecular docking

CompassDock 🧭

Navigating Future Drugs with CompassDock 🧭

The CompassDock framework is a comprehensive and accurate assessment approach for deep learning-based molecular docking. It evaluates key factors such as the physical and chemical properties, bioactivity favorability of ligands, strain energy, number of protein-ligand steric clashes, binding affinity, and protein-ligand interaction types.

Quickstart for CompassDock

conda create --name CompassDock python=3.11 -c conda-forge
conda install -c ostrokach-forge reduce
conda install -c conda-forge openbabel
conda install -c conda-forge datamol
pip install compassdock
pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.3.0+cu121.html
pip install "fair-esm @ git+https://github.com/asarigun/esm.git"
pip install "dllogger @ git+https://github.com/NVIDIA/dllogger.git"
pip install "openfold @ git+https://github.com/asarigun/openfold.git"

from compassdock import CompassDock

cd = CompassDock()

cd.recursive_compassdocking(
        protein_path = 'example/proteins/1a46_protein_processed.pdb',
        protein_sequence = None,
        ligand_description = 'CCCCC(NC(=O)CCC(=O)O)P(=O)(O)OC1=CC=CC=C1',
        complex_name = 'complex_1',
        molecule_name = 'molecule_1',
        out_dir = 'results',
        compass_all = False)

results = cd.results()

print(results)

More examples coming soon on Google Colab!

CompassDock 🧭 in Fine-Tuning Mode

For instructions on how to use Fine-Tuning Mode, please refer to the previous branch

Citation

Please cite the following paper if you use this code/repository in your research:

@article{sarigun2024compass,
  title={Compass: A Comprehensive Tool for Accurate and Efficient Molecular Docking in Inference and Fine-Tuning},
  author={Sarigun, Ahmet and Franke, Vedran and Akalin, Altuna},
  journal={arXiv preprint arXiv:2406.06841},
  year={2024}
}

Acknowledgements

We extend our deepest gratitude to the following teams for open-sourcing their valuable Repos:

• DiffDock Team (version 2023 & 2024),
• AA-score Team,
• PoseCheck Team