ApoDock is a modular docking paradigm that combines machine learning-driven conditional side-chain packing based on protein backbone and ligand information with traditional sampling methods to ensure physically realistic poses.
To run ApoDock, clone this GitHub repository and install Python.
- Operating System: Linux (Recommended)
conda env create -f environment.yaml
conda activate apodock
Then install gvp-gnn:
cd ApoDock_public
git clone https://github.com/drorlab/gvp-pytorch.git
cd gvp-pytorch
pip install -e .
First, you need to download the binary sampling software Gnina or Smina to docking_program
dir, then give execution permission:
chmod +x Gnina
chmod +x Smina.static
Then set the enviroment path:
export PATH="$PATH:/your/path/to/ApoDock_public/docking_program"
A demo:
python docking.py --protein ./demo/1a0q/1a0q_protein.pdb --ligand ./demo/1a0q/1a0q_ligand.sdf --ref_lig ./demo/1a0q/1a0q_ligand.sdf --packing
Use .CSV
file for docking:
python docking.py --csv docking_list.csv --packing
Output example will in defalut dir docking_results
, you can use --out_dir
option to determine the output position.
This work draws upon code from ProteinMPNN, OpenFold, RTMscore, and PIPPack, and we would like to thank them for their excellent contributions. these studies are important and interesting.