VMD-L Mailing List

From: Olya Kravchenko (ovkrav_at_gmail.com)
Date: Fri May 05 2017 - 13:27:39 CDT

Got it! Thank you very much for your input.

Olga

On Fri, May 5, 2017 at 2:26 PM, Radak, Brian K <bradak_at_anl.gov> wrote:

> My only warning was that the PSF needs to be rebuilt from scratch, rather
> than just patched. I don't know how mutator works, but I hope it does the
> former. However, now that I think about it, if mutator can handle
> completely different residue types, then it almost certainly rebuilds from
> scratch and my warning is completely unnecessary.
>
> I've been implementing something like this completely inside psfgen
> (without the benefit of the swanky tools inside VMD) and got tripped up by
> the situation that I described.
>
>
> Brian Radak
> Postdoctoral Appointee
> Leadership Computing Facility
> Argonne National Laboratory
>
> 9700 South Cass Avenue, Bldg. 240
> Argonne, IL 60439-4854
> (630) 252-8643
> brian.radak_at_anl.gov
> ------------------------------
> *From:* Olya Kravchenko [ovkrav_at_gmail.com]
> *Sent:* Friday, May 05, 2017 1:12 PM
> *To:* Radak, Brian K
> *Cc:* Vermaas, Joshua; vmd-l_at_ks.uiuc.edu
>
> *Subject:* Re: vmd-l: substitute one residue without rebuilding the
> structure?
>
> Hi Brian,
>
> psf is rebuild automatically when mutator is utilized. Or did you mean it
> is better to rebuild the molecule from scratch? I'd rather avoid rebuilding
> the molecule because I want to save computation time for equilibration, and
> because I want to have as little difference as possible with the original
> system.
>
> Thanks,
>
> Olga
>
> On Fri, May 5, 2017 at 9:54 AM, Radak, Brian K <bradak_at_anl.gov> wrote:
>
>> I don't have a lot of experience with mutator, but be careful here - the
>> structure of improper torsions is totally different for all HIS variants.
>> The bonds need to be totally removed and rebuilt.
>>
>> Personally I think the best bet is to completely rebuild the PSF after
>> changing the resname.
>>
>> But like I said, I have little idea as to what mutator actually does
>> underneath.
>>
>> Cheers,
>> Brian
>>
>> Brian Radak
>> Postdoctoral Appointee
>> Leadership Computing Facility
>> Argonne National Laboratory
>>
>> 9700 South Cass Avenue, Bldg. 240
>> Argonne, IL 60439-4854
>> (630) 252-8643
>> brian.radak_at_anl.gov
>>
>> ________________________________________
>> From: owner-vmd-l_at_ks.uiuc.edu [owner-vmd-l_at_ks.uiuc.edu] on behalf of
>> Olya Kravchenko [ovkrav_at_gmail.com]
>> Sent: Thursday, May 04, 2017 10:00 PM
>> To: Vermaas, Joshua
>> Cc: vmd-l_at_ks.uiuc.edu
>> Subject: Re: vmd-l: substitute one residue without rebuilding the
>> structure?
>>
>> Hi Joshua,
>>
>> thank you for your help. I tried mutator on my macbook and it works well,
>> it only fails on ubuntu. Indeed it looks as if it can't find topology
>> files, probably because I have an older version of vmd installed there.
>>
>> Olga
>>
>> On Thu, May 4, 2017 at 7:26 PM, Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov
>> <mailto:Joshua.Vermaas_at_nrel.gov>> wrote:
>> Hi Olga,
>>
>> Mutator ought to work. What is at the head of the error messages? In
>> principle mutator for your case is super straightforward, since the only
>> atomic coordinate that needs to be guessed is the proton on CE (HE2 I
>> think?), but the errors you are showing indicate that the plugin didn't
>> load the top_all36_prot.rtf topology file like it should have, so mutator
>> is bailing out on you. If you don't feel like diagnosing the plugin, you
>> can of course just do it by hand, using something like this:
>>
>> package require psfgen
>> mol load psf old.psf pdb old.pdb ;# Alternatively, use namdbin
>> old.restart.coor of a post-equilibration system
>> set oldseg [atomselect top "segment P1"]
>> set oldsegHSD [atomselect top "segment P1 and resid 1234"]
>> $oldsegHSD set resname HSE
>> $oldseg writepdb renamed.pdb
>> topology top_all36_prot.rtf
>> segment P1 {
>> pdb renamed.pdb
>> }
>> coordpdb renamed.pdb P1
>> regenerate angles dihedrals
>> guesscoord
>> writepsf p1.psf
>> writepdb p1.pdb
>>
>> Unfortunately rebuilding the psf is just unavoidable, since the bonds,
>> angles, and dihedrals change since the proton moved.
>>
>> -Josh
>>
>> On 05/04/2017 05:01 PM, Olya Kravchenko wrote:
>> Hi all,
>>
>> this must be something simple, but I couldn't come up with a good
>> solution. I need to change only one histidine residue from HSD to HSE and
>> run my ABF simulation to compare the outcome. The structure is already well
>> equilibrated, so I am wondering if there is a way to do this substitution
>> without rebuilding this system from scratch (to avoid long equilibration of
>> the whole molecule)? Can I do substitution through atom selection somehow
>> and just switch one HSD for HSE?
>>
>> I tried using mutator plugin, but it gives me errors of this sort:
>>
>> ...
>> ...
>> psfgen) unknown residue type THR
>> psfgen) unknown residue type LEU
>> psfgen) unknown residue type GLY
>> psfgen) extracted 172 residues from pdb file
>> psfgen) Info: generating structure...psfgen) unknown residue type THR
>> failed!
>>
>> Can mutator plugin be used for my goal and what am I doing wrong?
>>
>> I will appreciate your input!
>>
>> Thanks,
>>
>> Olga
>>
>>
>>
>