Openbabel

Overview

Open Babel is a chemical toolbox designed to search, convert, analyze, or store data from molecular modelling, chemistry, solid-state materials, biochemistry, or related areas.

Version 2.4.1 is available. It was compiled with the Intel compiler version 17.0.7 and python 2.7.

Please note that the GUI is not installed for either version and so Open Babel should be used for batch work only.

Restrictions on use

There are no restrictions on accessing the software. The source code is release under the GPL v2 license.

Set up procedure

To access the software you must first load the required modulefile:

Version 2.4.1

module load apps/intel-17.0/obabel/2.4.1        (1-core) jobs

OR

module load apps/intel-17.0/obabel/2.4.1-omp    # Multithreaded MMFF94 force-field code

Running the application

Please do not run Open Babel on the login node. Jobs should be submitted to the compute nodes via batch.

Serial batch job submission

Open Babel is mainly a serial code and so should only be run with a serial batch scripts (no PE flag). If you wish to run many similar jobs please see the instructions on job arrays for the correct way to do this.

Make sure you have the modulefile loaded then create a batch submission script, for example:

#!/bin/bash --login
#$ -cwd             # Job will run from the current directory
module load apps/intel-17.0/obabel/2.4.1
babel infile outfile

Submit the jobscript using:

qsub scriptname

where scriptname is the name of your jobscript.

Parallel batch job submission

The MMFF94 force-field code is multicore aware in the 2.4.1-omp version. If you are running OpenBabel and specifying the following flags on the command-line:

-ff MMFF94

then your job may benefit from being run with multiple cores.

Please note: If you are not using the MMFF94 code do not use the -pe line in your jobscript – use a serial jobscript. Requesting multiple cores without using the MMFF94 code will simply waste cores because your code will use only one core.

For example:

Make sure you have the 2.4.1 or 2.3.2-omp modulefile loaded then create a batch submission script, for example:

#!/bin/bash --login
#$ -cwd             # Job will run from the current directory
#$ -pe smp.pe 4     # Job will use 4 cores (max 32 allowed)

# Must set this to use the correct number of cores
export OMP_NUM_THREADS=$NSLOTS

# Must be using the MMFF94 force-field code
obminimize -ff MMFF94 other args molfile