Thanks for the response - Mostly seem to be in business now (sans eefx pot).
One issue we are having is that the energy summaries are not printing with the structures and the violation analysis is not working. Is this something to do with using the atcun.par file? It doesn't seem to be related to using the prePot and the script without the ATCUN additions does print the energy and violation summaries as expected. Lisa -----Original Message----- From: xplor-nih-boun...@dcb.cit.nih.gov [mailto:xplor-nih-boun...@dcb.cit.nih.gov] On Behalf Of Charles Schwieters Sent: Tuesday, October 18, 2016 12:45 PM To: tuttl...@u.washington.edu Cc: xplor-nih@dcb.cit.nih.gov Subject: Re: [Xplor-nih] ensemble reifnement with ATCUN PRE data Hello Lisa-- > > I'm attempting to refine a structure against ATCUN PRE R2 relaxation > data, but am hitting some errors. > > We've successfully added the ATCUN motif to the protein using the > eginput/PSF_generation/tags/atcun_motif/addATCUN.py script > > We've modified an ensemble script (with eefx now working, thank you), > starting from the newly generated ATCUN-protein.pdb and adding the > prePot potential My colleague Guillermo Bermejo pointed out to be that you must be careful using actun top/par values with eefx- they are not compatible. Things may work but there are no guarantees. > rampedParams.append( StaticRamp(setupSBMFmode( pre )) ) > highTempParams.append( StaticRamp(setupSBmode( pre )) ) I think you'll want SBmode throughout, at least initially. > > --- > > This will run fine (if not doing a great job satisfying the PRE data), but is giving an error during writeStructure: > "Error calculating structure 0: 'PREPot' object has no attribute '__getitem__' > Warning: fitting EnsembleSimulation structures" > > The bottom of the output file: > > ensemble-atcun.py(384): loopInfo.writeStructure(potList) > Traceback (most recent call last): This is a simple bug, which is corrected in the attached version of prePotTools.py which should be placed in the python subdirectory of your Xplor-NIH 2.43 distribution. > > So - do we need to explicitly include the atcun.par file in our > refine runs as is done int he addATCUN.py? Or is the error coming > from something else? You should include atcun.par, but all bets are off when used with eefx. You might try with standard top/par values first. best regards-- Charles _______________________________________________ Xplor-nih mailing list Xplor-nih@dcb.cit.nih.gov https://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
