VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Fri Nov 18 2005 - 09:59:16 CST

Vlad,
  STRIDE is fed it's data directly from VMD, not from the original file.
I thought about it some more this morning, and now I believe that your
problem has nothing to do with the bonds at all, because the information
that VMD sends to STRIDE is done by writing a new PDB file containing the
structure from VMD's memory, executing STRIDE on it, and getting the results
back. I wasn't thinking clearly when I suggested toggling that bond on/off.
Toggling the bond on/off clearly won't make a difference since bonding
information isn't provided in the PDB file that VMD generates before it
calls STRIDE. Given this, its far more likely that STRIDE is sensitive to
the residue names and that the PDB file you tried previously had a residue
name recognized by STRIDE, and the parm file did not.
Can you compare the residue names that were assigned
in your PDB version versus what you get when you load directly from the
AMBER parm file? I suspect STRIDE may be lost due to the residue naming...

Let me know,
  John Stone
  vmd_at_ks.uiuc.edu

On Fri, Nov 18, 2005 at 11:57:08AM +0100, Vlad Cojocaru wrote:
> Dear John,
> Thanks a lot for the suggestions .. However it seems that they do not
> solve the problem since the bond is present in the amber topology file,
> no matter if I graphically remove it in VMD, and I guess STRIDE tries to
> calculate the SS from the original file (correct me if I am wrong). Of
> course a provisional solution could be to load a pdb file instead of the
> topology file and then load the trajectory onto this pdb file .....
> However, it would be nice if this problem will be eventually solved...
> Best wishes
> vlad
>
> John Stone wrote:
>
> >Vlad,
> > How about temporarily taking out the bond that confuses stride using the
> >Mouse->Add/Remove Bonds feature. Once removed, you could then use sscache
> >to get stride to build 2ndary struct for all of your timesteps, or if you just
> >need one good one, just call 'mol ssrecalc top'. Once you've got STRIDE
> >doing what you need, you can put the bond back, and things should be ok then.
> >This is obviously an annoying hack, but until someone comes up with a better
> >free/open source secondary structure determination package, I don't have
> >any better suggestions. That said, if anyone knows of something better,
> >I'm all ears.
> >
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> >On Wed, Nov 16, 2005 at 12:14:47PM +0100, Vlad Cojocaru wrote:
> >
> >
> >>Dear vmd users,
> >> In my protein I have a metal ion that has five coordinations, one of
> >>them being a CYS residue. For AMBER, I built a new residue RES by
> >>creating a bond between the S atom of CYS and the metal ion. If I load
> >>into vmd the AMBER topology together with the AMBER coordinates files,
> >>the bond S-Me is drawn but STRIDE fails to calculate the secondary
> >>structure complaining that there are too many atoms in the unusual
> >>residue. If I load a pdb file of the protein, the S-Me bond is not drawn
> >>and STRIDE calculates the secondary struucture properly. This is not
> >>really nice because I am not able to create NewCartoon representations
> >>for my molecule unless I convert all amber trajectories to multiple pdb
> >>files (which of course I dont want).
> >> The question is ... Is there a way I can make STRIDE recognize my
> >>unusual residue as regular protein residue so that I can represent the
> >>secondary structure of the protein without converting AMBER trajectories
> >>to multiple pdb files?
> >>
> >>
> >>Thank you very much in advance
> >>
> >>Best wishes
> >>vlad
> >>
> >>
> >> Dr. Vlad Cojocaru
> >> EML Research gGmbH
> >> Molecular and Cellular Modeling Group
> >> Villa Bosch
> >> Schloss-Wolfsbrunnenweg 33
> >> 69118 Heidelberg
> >> Germany
> >> Phone: +49-6221-533266
> >> Fax: +49-6221-533298
> >> e-mail: Vlad.Cojocaru_at_eml-r.villa-bosch.de
> >> <mailto:Vlad.Cojocaru_at_eml-r.villa-bosch.de>
> >> Additional info: my home page
> >> <http://projects.villa-bosch.de/mcm/people/cojocaru/>
> >>
> >>
> >>
> >>
> >
> >
> >
>
> --
>
>
> Dr. Vlad Cojocaru
> EML Research gGmbH
> Molecular and Cellular Modeling Group
> Villa Bosch
> Schloss-Wolfsbrunnenweg 33
> 69118 Heidelberg
> Germany
> Phone: +49-6221-533266
> Fax: +49-6221-533298
> e-mail: Vlad.Cojocaru_at_eml-r.villa-bosch.de
> <mailto:Vlad.Cojocaru_at_eml-r.villa-bosch.de>
> Additional info: my home page
> <http://projects.villa-bosch.de/mcm/people/cojocaru/>
>
> 

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078