[3dem] IHRSR hseach and Relion

Qiu-Xing Jiang Qiu-Xing.Jiang at UTSouthwestern.edu
Mon Apr 14 09:58:09 PDT 2014


Dear all,
I thought that the following technical details may be of interest for those using IHRSR or Relion.
For Hernando’s question, we modified the pointer position in the fortran files in order to run them with the right output in a new HPC Linux cluster. Hernando tested our modified files to be working in his new system. He compared the results with what coming out of his earlier version of  hsearch and found that the numbers varied less than 5%. If you would like to try these files, please contact me or Hernando.
To run Relion across the inifiband interface in a new HPC cluster, we used the hidden option "--dont_combine_weights_via_**disc” to accelerate the refinement. There was a warning message complaining that it was not a valid option, but Relion ran much faster. Sjors wanted this info to be shared.
Thanks.
Qiu-Xing

From: Hernando J Sosa <hernando.sosa at einstein.yu.edu<mailto:hernando.sosa at einstein.yu.edu>>
Date: Thursday, April 10, 2014 at 9:00 PM
To: "3dem at ncmir.ucsd.edu<mailto:3dem at ncmir.ucsd.edu>" <3dem at ncmir.ucsd.edu<mailto:3dem at ncmir.ucsd.edu>>
Subject: [3dem] IHRSR hseach

I  wonder if anybody has encountered this problem an may know how to fix it:

When running the program hsearch (or hsearch_lorents or hsearch_lorentz.latest) from  the IHRSR routines  in a  new linux box (relatively recently installed OS)    the program seems to run fine but  at the end does not write the results  to the doc text file and gives the following error:

At line 146 of file hsearch_lorentz.latest.F (unit = 11, file = 'ihrsr_symdoc.txt')
Fortran runtime error: Sequential READ or WRITE not allowed after EOF marker, possibly use REWIND or BACKSPACE

The same program with the same file runs  fine in another computer. I  recompiled it but the problem persists.

Any idea?

Thanks

Hernando

________________________________

UT Southwestern Medical Center
The future of medicine, today.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.ncmir.ucsd.edu/mailman/private/3dem/attachments/20140414/1ab8fc5f/attachment.html>


More information about the 3dem mailing list