Page 1 of 1

Using a super computer

PostPosted: Tue Jun 21, 2016 6:27 pm
by Tanjay
Hi

I would like to use a super computer to run every combination i have between proteins/ligand but for that i might not have access to the GUI. Is there any way to produce an input file in which i write all my parameters that PyRx will be able to read and use in order to start? Basically, i want to run PyRx using an input file from within a batch script.

Also, do you have any example of a batch script that would run PyRx using the selected input file.

Thank you,

Tanjay

Re: Using a super computer

PostPosted: Tue Jun 21, 2016 8:39 pm
by sargis
Hi,

PyRx is using qsub command to submit jobs to the cluster. If you don't have access to the GUI, I suggest to contact the sysadmin to have a GUI access on the head node. If this is somehow is not possible, then you can run PyRx from another computer that has GUI. You'll need to have a qsub command in your path. You can make, for instance, a symbolic link to ls command and name it qsub. PyRx will then have an option to make all jobs files, which you can later transfer to cluster.

The following links provide excellent information on how do I submit a large number of very similar jobs and how to run shell loop over set of files
https://marylou.byu.edu/wiki/index.php? ... ar+jobs%3F
http://www.cyberciti.biz/faq/bash-loop-over-file/

An example of a batch script that would run using the jobs files created by PyRx would look something like:
Code: Select all
cd ~/.PyRx_workspace/Macromolecules/your_pdb
for f in PyRxVinaJob*
do
   qsub $f
done