First, I have a hopefully easy question. I can't seem to figure it out although feel like I should be able to. Is there a way to run a phenix.refine job in the background? I tried using the & like I usually do and it didn't work. It would be nice to start a job and go home without leaving myself logged into the computer. I didn't see anything in the documentation which makes me think I'm being dull. Second, a technical question. When phenix.refine does ordered solvent, does it use detwinned maps if you've put in the appropriate twin operator and indicated it was twinned? When I've tried running phenix.refine with my perfect twin, it doesn't seem to find waters like I can using the CNS twinned water picking script. Third, is it possible to run phenix.refine in parallel at all. My structure is annoying large and the parallelization of FFT in CNS seems to help speed things up abit. Thanks, Mary X. Fitzgerald Postdoctoral Associate Arnold Lab
On Feb 21, 2008, at 12:23 PM, Mary Fitzgerald wrote:
First, I have a hopefully easy question. I can't seem to figure it out although feel like I should be able to. Is there a way to run a phenix.refine job in the background? I tried using the & like I usually do and it didn't work. It would be nice to start a job and go home without leaving myself logged into the computer. I didn't see anything in the documentation which makes me think I'm being dull.
screen is a personal favorite. screen phenix.refine file.sca model.pdb then CTRL A-D to disconnect, logout. You could use screen - r to resume, but by the time you login again the job will probably be done. screen is installed by default on OS X. FR --------------------------------------------- Francis Reyes M.Sc. 215 UCB University of Colorado at Boulder gpg --keyserver pgp.mit.edu --recv-keys 67BA8D5D 8AE2 F2F4 90F7 9640 28BC 686F 78FD 6669 67BA 8D5D
The & should work actually. What makes you think it doesn't?
Regarding twinning, detwinned maps are used. There was a bug somewhere
that was fixed recently. I suggest you download the latest cci_apps
and try it with that version.
HTH
Peter
2008/2/21, Mary Fitzgerald
First, I have a hopefully easy question. I can't seem to figure it out although feel like I should be able to. Is there a way to run a phenix.refine job in the background? I tried using the & like I usually do and it didn't work. It would be nice to start a job and go home without leaving myself logged into the computer. I didn't see anything in the documentation which makes me think I'm being dull.
Second, a technical question. When phenix.refine does ordered solvent, does it use detwinned maps if you've put in the appropriate twin operator and indicated it was twinned? When I've tried running phenix.refine with my perfect twin, it doesn't seem to find waters like I can using the CNS twinned water picking script.
Third, is it possible to run phenix.refine in parallel at all. My structure is annoying large and the parallelization of FFT in CNS seems to help speed things up abit.
Thanks,
Mary X. Fitzgerald Postdoctoral Associate Arnold Lab
_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
Hi Mary,
First, I have a hopefully easy question. I can't seem to figure it out although feel like I should be able to. Is there a way to run a phenix.refine job in the background? I tried using the & like I usually do and it didn't work.
In this respect, phenix.refine is like any other Unix command. It will work if you direct both "stdout" and "stderr" to a file, e.g. in tcsh: phenix.refine model.pdb data.mtz >& logfile & I think newer bash versions also support the simple ">&" syntax. It is a good habit to redirect the output of all background jobs in this way. The --quiet option pointed out by others is a good idea, too. But to be sure a background command doesn't hang always use the redirection to a file.
Third, is it possible to run phenix.refine in parallel at all. My structure is annoying large and the parallelization of FFT in CNS seems to help speed things up abit.
No, sorry. We will address this at some point in the future. Ralf
Thanks for answering.
I must have been half asleep when I tried using & before in linux. I
remember it not working but I probably misskeyed something as it was the
first couple of time I ran phenix. It's working now.
I was using a cci application from January so I'll try the newest version
for the ordered solvent.
Thanks again,
Mary
On 2/21/08, Ralf W. Grosse-Kunstleve
Hi Mary,
First, I have a hopefully easy question. I can't seem to figure it out although feel like I should be able to. Is there a way to run a phenix.refine job in the background? I tried using the & like I usually do and it didn't work.
In this respect, phenix.refine is like any other Unix command. It will work if you direct both "stdout" and "stderr" to a file, e.g. in tcsh:
phenix.refine model.pdb data.mtz >& logfile &
I think newer bash versions also support the simple ">&" syntax. It is a good habit to redirect the output of all background jobs in this way. The --quiet option pointed out by others is a good idea, too. But to be sure a background command doesn't hang always use the redirection to a file.
Third, is it possible to run phenix.refine in parallel at all. My structure is annoying large and the parallelization of FFT in CNS seems to help speed things up abit.
No, sorry. We will address this at some point in the future.
Ralf _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb
participants (4)
-
Francis E Reyes -
Mary Fitzgerald -
Peter Zwart -
Ralf W. Grosse-Kunstleve