[gmx-developers] Fwd: Re: using replace/delete syntax of termini database for specbond

Thu Aug 26 08:13:14 CEST 2010

One additional thing would need to be changed for my proposal to work for
polymers with different linkages.

If I'm not mistaken, specbond is currently applied the following way:

As an example consider a residue X which has two linkages and thus 2 rules
apply. Let's assume the first rule would rename X to X1 and the 2nd rule
would rename X to XA. Even if the first rule is applied before the 2nd rule
then the 2nd rule still has to refer to X/X1 as X and not X1. Thus columns 1
and 4 always refer to the residue with its originally name given in the pdb
file and not by its new name given by a prior rule.

This behavior makes it impossible to rename a residue consecutive based on
two rules. Being able to rename in two steps would allow to use specbonds in
cases like lignin with different possible linkages. One could use one
specbond rule to rename X to X1 or X2 depending of the preceding linkage "1"
or "2" in the chain and then to X1A, X1B, X2A or X2B depending on the
following linkage A or B.

Thus we would need to either:
- Change the behavior that specbond uses the prior renamed residue name for
a following rule for columns 1-4, or
- Have additional columns which would allow to specify the name of a residue
based on the prior changed name of an earlier rule

 Roland

On Wed, Aug 25, 2010 at 12:17 PM, Roland Schulz <roland at utk.edu> wrote:

>
>
> On Wed, Aug 25, 2010 at 10:18 AM, Pär Bjelkmar <bjelkmar at cbr.su.se> wrote:
>
>>
>> It's something similar to what CHARMM does. They use a lot of such special
>> linkages to combine existing "residues" into new ones, all protonation
>> states for example but also some much more complicated things like
>> "building" the DPPC lipid from Palmitate (PALM) and "PCGL", using two
>> linkages (EST1 and EST2), called patches in CHARMM.
>>
>> Yes. It is similar to the CHARMM patches. But I tried to modify how
> patches work within CHARMM to a a format which matches the current syntax of
> terminal databases and specbond, with the hope to make it easy to
> implement.
>
>> And this is also, I guess(?), why you, Roland, used the trick with the
>> modified psfgen program (NAMD/VMD software) to build GROMACS topologies of
>> your lignin and cellulose molecules: it could build the molecules modularly
>> using the patches.
>>
>>  Yes. While cellulose is easy to describe with GROMACS topogies lignin is
> more difficult. Lignin has 4 different linkages and can be branched. Thus a
> residue can be modified at 2 or 3 sites by either one of the 4 linkages.
> Thus the monomer can be modified in 4^3+4^2 different ways (a few less
> because some are exclusive but other linkages also exist). Generating those
> ~80 versions of a monomer by hand is of course not very nice. Thus we
> decided to use the modified psfgen. Now we would like to try virtual sites
> with lignin and thus we look at this issue again.
>
>> Regarding the CHARMM port: this modularity is why it's kind-of hard to
>> port the CHARMM ff "completely".
>>
>> Would my proposed extension (using the tdb syntax similar to patches) help
> with porting the parts of the CHARMM ff which remains to be ported?
>
> Roland
>
>>
>> /Pär Bjelkmar
>>
>> Message: 1
>>
>> Date: Tue, 24 Aug 2010 19:44:07 -0400
>>
>> From: Roland Schulz <roland at utk.edu>
>>
>> Subject: [gmx-developers] using replace/delete syntax of termini
>>
>> database for specbond
>>
>> To: Discussion list for GROMACS development
>>
>> <gmx-developers at gromacs.org>
>>
>> Message-ID:
>>
>> <AANLkTim5=wX060wcv29XxuApPXdhEK8DLYnxg9dn23a_ at mail.gmail.com>
>>
>> Content-Type: text/plain; charset="iso-8859-1"
>>
>>
>> Hi,
>>
>>
>> it would be very nice if it would be possible to use the
>> replace/delete/add
>>
>> features of the termini (tdb) files also in combination with specbond.
>> This
>>
>> would make it simpler to specify some non-standard linkages.
>>
>>
>> As an example: Instead of fully specifying the CYS2 it would be nice to be
>>
>> able to say:
>>
>> [ CYS2 ]
>>
>> [ replace ]
>>
>>   CB    opls_214    12.011 0.0975
>>
>>   SG    opls_203   32.06500 -0.2175
>>
>> [ delete ]
>>
>>   HG
>>
>>
>> For this example the saving would be very limited. But with a large number
>>
>> of possible linkages this would give a considerable saving in specifying
>> the
>>
>> topology for possible combinations of linkages. The residue used as basis
>>
>> (CYS) for the modified residue (CYS2) could come from the beginning of the
>>
>> name or could be specified explicitly. This modified residue could be used
>>
>> by a specbond line (which wouldn't change from the current line using
>>
>> CYS2). It seems this feature  could be implemented by a very limited patch
>>
>> because it would use only features already implemented for termini. But I
>>
>> don't understand the termini code so I can't judge this myself. Also I'm
>> not
>>
>> sure whether it might be a problem with the charge groups.
>>
>>
>> Would this be something easy and useful to add to pdb2gmx?
>>
>>
>> Roland
>>
>>
>> --
>>
>> ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
>>
>> 865-241-1537, ORNL PO BOX 2008 MS6309
>>
>> -------------- next part --------------
>>
>> An HTML attachment was scrubbed...
>>
>> URL:
>> http://lists.gromacs.org/pipermail/gmx-developers/attachments/20100824/e1017c7b/attachment-0001.html
>>
>>
>> ------------------------------
>>
>>
>> --
>>
>> gmx-developers mailing list
>>
>> gmx-developers at gromacs.org
>>
>> http://lists.gromacs.org/mailman/listinfo/gmx-developers
>>
>>
>>
>> End of gmx-developers Digest, Vol 76, Issue 15
>>
>> **********************************************
>>
>>
>>
>>
>>
>> --
>>
>> gmx-developers mailing list
>> gmx-developers at gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-developers
>> Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-developers-request at gromacs.org.
>>
>
>
>
> --
> ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
> 865-241-1537, ORNL PO BOX 2008 MS6309
>

-- 
ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
865-241-1537, ORNL PO BOX 2008 MS6309
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20100826/e9221779/attachment.html>