Hello Charles,
While I was going through example scripts, in eginput/gb3_ensemble and
comparing against eginput/gb3_rdc [single structure refinements], I found
some thing that looks weird.
In eginput/gb3_ensemble:
dyn = ivm.IVM(esim) #manipulator for T-A dynamics
dynx = ivm.IVM(esim) #manipulator for Cartesian dynamics
but dynx was never used as a cartesian dynamics [no call to
protocol.cartesianTopology(dynx).]
Is there a reason to do that? How to perform a cartesian minimization when
running an ensemble simulation?
I have tried to mention,
protocol.torsionTopology(dyn)
protocol.cartesianTopology(dynx)
but the resulting refinements should some bigger R-factors than ones that
did not.
Thanks for your time and advice,
santhosh
On Tue, Aug 27, 2013 at 1:38 PM, santhu kumar <mesan...@gmail.com> wrote:
> Hello Charles,
>
> Thanks. That makes sense and it works. :-)
>
> Good day,
> Santhosh
>
>
> On Tue, Aug 27, 2013 at 12:15 PM, Charles Schwieters <
> char...@schwieters.org> wrote:
>
>>
>> Hello Santhos--
>>
>> >
>> > I was able to successfully load the pdb files using both loadPDB and
>> > initCoords [i generated a psf file using pdb2psf and since the psf
>> would be
>> > the same for all the ensemble simulation members I think thats OK].
>> >
>> > But I am not really sure if its doing the right thing. Just to be clear
>> of
>> > the question:
>> > I want different ensemble members to have different starting
>> structures.
>> > Here is the snippet of the code, my reason for confusion in comments :
>> >
>> > protocol.initRandomSeed()
>> > protocol.initParams("protein")
>> > ensembleSize = 2
>> > pdbFiles = ["model0.pdb","model1.pdb"] # these two files should be the
>> > starting structures
>> >
>> > esim = EnsembleSimulation("ensemble",ensembleSize)
>> > memberIndex = esim.member().memberIndex() # this makes memberIndex = 0
>> > protocol.initStruct('model.psf') # the default psf file
>> > protocol.initCoords(pdbFiles) #
>>
>> close: please instead use
>>
>> protocol.initCoords( pdbFiles[memberIndex%2] )
>>
>> >
>> > The same code as Charles has wrote :
>> > protocol.initCoords(pdbFiles[memberIndex%2]) # using 2 structures,
>> > memberIndex%2 would always be zero as memberIndex =0, means the code
>> loads
>> > only one file.
>>
>> No. memberIndex is 0 or 1 for a 2-membered ensemble. To better see
>> what is going on, you can enable output for those having non-zero
>> memberIndex by adding the flag
>>
>> -thread_filename thread.out
>>
>> to the xplor command invocation. You will then have files for each
>> non-zero ensemble member ion your directory.
>>
>> best regards--
>> Charles
>>
>
>
_______________________________________________
Xplor-nih mailing list
Xplor-nih@cake.cit.nih.gov
http://cake.cit.nih.gov/mailman/listinfo/xplor-nih
