Skip to content

AutoDock GPU


The AutoDock-GPU Suite is a growing collection of methods for computational docking and virtual screening, for use in structure-based drug discovery and exploration of the basic mechanisms of biomolecular structure and function. The source code is available at github page.

It is highly recommended to run AutoDock GPU within the container using Singularity utility.


Following versions of AutoDock GPU are currently available:

  • Runtime dependencies:
    • singularity

You can load dependencies as modules by following command:

   module load singularity

User guide

You can find the software documentation and user guide at official website and for running in singularity at nvidia catalog.

Example run script

You can copy and modify this script to and submit job to a gpu node by command sbatch Values for number of GPUs, ntasks and CPUs per task have been tested and optimized for best performance on Devana cluster.

#SBATCH -J "AD_GPU"     # name of job in SLURM
#SBATCH --account=<project> # project number
#SBATCH --partition=ngpu    # selected partition
#SBATCH --gres=gpu:1        # runs on one gpu
#SBATCH --ntasks=1      # number of parallel tasks
#SBATCH --cpus-per-task=8   # number of cpus per task
#SBATCH --time=hh:mm:ss     # time limit for a job
#SBATCH -o stdout.%J.out    # standard output
#SBATCH -e stderr.%J.out    # error output

# Modules
module load singularity

# Shortcuts

# Execute script within the container 
singularity exec --nv $Container_loc/autodock_gpu_zinc_latest.sif "$Script_loc/"

Second script that is started within the container loops over desired directory, can be modified to specify targets or loop over list of ligands (see section AutoDock Vina).


# Shortcuts

# Loop over compounds
Ligands=$(ls $Ligands_location/*.pdbqt)

for ligand in $Ligands; do
    #   autodock_gpu_128wi -lfile $ligand -ffile $Proteins_location/6WQF_clean.maps.fld -nrun 50 -resnam $Outputs/$ligand.dlg
    autodock_gpu_128wi -lfile $ligand -ffile $Proteins_location/6WQF_clean.maps.fld -nrun 50
Created by: Marek Štekláč