DiffDockHPC

DiffDockHPC is a fork of DiffDock, which adds support to run DiffDock on HPC systems using Singularity and Slurm.
DiffDockHPC has been developed to be part of a consensus docking protocol: ESSENCE-Dock.
For more details about DiffDock itself, we refer to the DiffDock Github and the Paper on arXiv. DiffDockHPC current version matches to DiffDock v1.1 (DiffDock-L).
Note: If you update from DiffDockHPC v1.0, it is highly recommended to perform a clean install.

DiffDockHPC is also available for the original DiffDock v1.0 implementation. This version was used in the original ESSENCE-Dock paper. In case you want to work with DiffDockHPC using the DiffDock 1.0 implementation, you can clone the project, and use git checkout DiffDockHPCv1.0.

Requirements:

Singularity
Slurm (There is a --no_slurm mode, but using Slurm is highly recommended)

Installation instructions:

Clone the repository and navigate to it

git clone https://github.com/Jnelen/DiffDockHPC

cd DiffDockHPC

Run a test example to automatically download the Singularity image (~3 GB) and to generate the necessary cache look-up tables for SO(2) and SO(3) distributions. (This only needs to happen once and usually takes around 15 minutes).
The --no_slurm flag is optional here, but makes it easier to track the progress.
```
python inferenceVS.py -p data/1a0q/1a0q_protein_processed.pdb -l data/1a0q/ -out TEST -j 1 --no_slurm
```
Or if you have access to a GPU, you can also add the -gpu tag like this:
```
python inferenceVS.py -p data/1a0q/1a0q_protein_processed.pdb -l data/1a0q/ -out TEST -j 1 -gpu --no_slurm
```

You can also download the Singularity image manually:

wget --no-check-certificate -r "https://drive.usercontent.google.com/download?id=1TsbuhNWA74AHfIbKV5uh2lmEnD99VlCD&confirm=t" -O singularity/DiffDockHPC.sif

alternatively, you can build the singularity image yourself using:

singularity build singularity/DiffDockHPC.sif singularity/DiffDockHPC.def

Options

The main file to use is inferenceVS.py. It has the following options/flags:

-p, -r, --protein_path: Path to the protein/receptor .pdb file.
-l, --ligand: The path to the directory of (separate) mol2/sdf ligand files.
--protein_ligand_csv: The path to a protein_ligand_csv file. Format and header should be like the following: complex_name,protein_path,ligand_description.
-o, --out, --out_dir: Directory where the output structures will be saved to.
-j, --jobs: Number of jobs to use.
-qu, --queue: On which node to launch the slurm jobs. The default value is the default queue for the user. Might need to be specified if there is no default queue configured.
-m, --mem: How much memory to use for each job. The default value is 4GB.
-gpu, --gpu: Use GPU resources. This will accelerate docking calculations if a compatible GPU is available.
-c, --cores: How many cores to use for each job. The default value is 1 when used with the GPU option enabled, otherwise it defaults to 4 cores.
-n, --num_outputs: How many structures to output per compound. The default value is 1.
--remove_hs: Remove the hydrogens in the final output structures.
--no_slurm: Don't use slurm to handle the resources. This will run all samples on 1 GPU. Other Slurm arguments such as the amount memory, time limit, ... will also be ignored. The amount of CPU cores will still be set.
--config: Path to the config file you want to use. Defaults to default_inference_args.yaml
-h, --help: Show the help message and exit.

License

MIT