Using Autodock on Biowulf

Status
About
Hardware
Applications
Batch queues
Disk storage

MPI
Performance
New Users
User Guide
Documentation
Research
Photos

Autodock on Biowulf

Autodock is a suite of automated docking tools. It is designed to predict how small molecules, such as substrates or drug candidates, bind to a receptor of known 3D structure. Autodock was developed at the Scripps Research Institute in San Diego.

Autodock v 3.0.5 is installed on the Biowulf cluster in /usr/local/autodock/ The executables addsol, atmtobnd, autodock3, autogrid3, autotors, makelaunch, protonate are available in /usr/local/autodock/dist305/bin. The main executable autodock3 is also available in /usr/local/bin in the default path of every user.

Autodock on Biowulf is intended for large-scale Autodock processing. Since the program is single-threaded, a single job cannot be parallelized over multiple processors. The best reason to run Autodock on the Biowulf cluster would be to run large numbers of Autodock jobs with different parameters or input files simultaneously.

It is not recommended that the Autodock-related graphics tools be run on the Biowulf cluster.

Autodock website.
Autodock v3 User Guide
Autodock tutorials.

Submitting a single Autodock job

Set up a job script along the following lines:

#!/bin/bash
#
#this file is Myjob.sh
#
#PBS -N Autodock
#PBS -m be
#PBS -k oe
#

cd $PBS_O_WORKDIR
autodock3 -p myfile.dpf -l myfile.dlg

Submit this job:

qsub -l nodes=1 Myjob.sh

Submitting a swarm of Autodock jobs

Set up a swarm command file with a single line for each Autodock command. e.g.

--------file myswarm.cmd--------------- autodock3 -p /data/user/mydir/lig1.macro.dpf -l /data/user/mydir/llig1.log autodock3 -p /data/user/mydir/llig2.macro.dpf -l /data/user/mydir/llig2.log autodock3 -p /data/user/mydir/llig3.macro.dpf -l /data/user/mydir/llig3.log autodock3 -p /data/user/mydir/llig4.macro.dpf -l /data/user/mydir/llig4.log [...]

Submit this swarm of jobs with a command like:

swarm -f myswarm.cmd

See the swarm web page for more information about swarm options.

It may be more convenient to separate the files for each autodock run into individual directories. In this case, the swarm command file could be set up as follows:

--------file myswarm.cmd--------------- cd /data/user/mydir1; autodock3 -p lig1.macro.dpf -l llig1.log cd /data/user/mydir2; autodock3 -p lig2.macro.dpf -l llig2.log cd /data/user/mydir3; autodock3 -p lig3.macro.dpf -l llig3.log

This document is available as http://biowulf.nih.gov/apps/autodock.html
Biowulf home page | Helix Systems | NIH