sreflex for neutrons. "ERROR log file does not exist"

Linear (OLIGOMER), and non-linear (MIXTURE) analysis, singular value decomposition (SVDPLOT), addition of missing fragments (BUNCH, CORAL), analysis of flexible systems (EOM/RANCH & GAJOE), flexible refinement of high-resolution models (SREFLEX)
Post Reply
Message
Author
Anne M
Member
Posts: 9
Joined: 2012.06.05 16:46

sreflex for neutrons. "ERROR log file does not exist"

#1 Post by Anne M » 2017.12.13 13:08

Dear Atsas team/Alejandro,
I am running sreflex on SANS data. My last trial almost worked but eventually failed with the strange error:

TIMED OUT: "cryson" ("-nosmy", "-p", "tmp\cp3\cp3", "tmp\cp3\cp3.pdb",
"tmp\cp3\cp3.dat")

after: 3600 seconds

ERROR -- FILE tmp\cp3\cp3.log DOES NOT EXIST

This is true that cp3.log does not exist. But I do not understand why
cryson did not make it. And it did not wait for 3600 seconds... It did
not make cp0.log and cp2.log either, but did not complain at this
stage... Can you explain, please?

On a more general point of view, I would need info on how sreflex works with neutron data (absolute intensity?, smearing?, H-exchange...):
From the manual, it seems that sreflex does not read the resolution column and you cannot pass the -res resolution file to cryson. Is it true? What does the --neutron option do? For neutrons, hydrogens are very important. Should the pdb file contain hydrogens if --neutron option is choosen? I can help in writing a neutron appendix to the manual if you like...

Thanks a lot.

sasha
Active member
Posts: 46
Joined: 2014.03.05 17:56

Re: sreflex for neutrons. "ERROR log file does not exist"

#2 Post by sasha » 2017.12.14 10:33

Dear Anne,

thanks for reporting this. I can't fully test it right now, but it looks like CRYSON is dying silently, probably related to RAM usage issues of the .bss segment in parallel threads. This seems particularly troublesome in Windows.
Please try running SREFLEX again with the extra argument: " -t 1 ", this will set the number of threads to 1 and probably solve the issue for now. Using a machine with more available RAM or another operating system may help as well.

I'm very glad there's interest in using SREFLEX with neutron data, but I clearly need to add the options you describe, as currently no smearing is done. Your input will be more than welcome, I'll contact you once I'm back in the lab.

Best,

sasha

Post Reply