VMD-L Mailing List

From: Paul R Brenner (pbrenne1_at_nd.edu)
Date: Tue Feb 01 2005 - 21:47:38 CST

Thanks again for the reply Jan. I'm guessing the VMD authors decided to export
only the atom indexs as they are the only items officially indexed in the pdb
or psf. The only bookeeping required is to make sure you realize that VMD
starts the atom indexing with zero. Providing indexs for the dihedrals based
on the relational informatin in the psf is probably so less "standardized" that
they felt such indices could easily lead to trouble when comunicated accross
platforms.

After reading your first response it dawned on me that with a little additional
code I can use the atom indices to determine the backbone (or alpha helix, beta
sheet, etc...) dihedrals within my MD software (I use Protomol).

Regards,
Paul

Quoting Jan Saam <saam_at_charite.de>:

> Paul,
>
> AFAIK, VMD keeps a list of dihedrals internally, but this list contains
> all possible dihedrals, not only the protein backbone diheds. (Just like
> the ones that are defined in the psf file.) Unfortunately, there is no
> tcl procedure to access this info.
>
> Maybe this is a good feature request, I could use such a function, too.
> John! Did you read this?
>
> If you just want the backbone diheds can build your own list by
> selecting each resid in each segid and determine the dihedrals respectively.
>
> Jan
>
> Paul R Brenner wrote:
> > Jan,
> >
> > Thanks for the reply. I am specifically interested in an index for the
> dihedral
> > itself (not an index of the atoms which make up the dihedral). I am hoping
> that
> > VMD has some way of indexing the dihedrals for a given set of atoms and can
> > subsequently output them.
> >
> > Paul
> >
> > Quoting Jan Saam <saam_at_charite.de>:
> >
> >
> >>Paul,
> >>
> >>I'm not sure, if I di understand you correctly, but I think the answer
> >>is quite easy:
> >>If you have an atomselection $sel then you can retrieve the indexes
> >>either with
> >>set myindexes [$sel list]
> >>or
> >>set myindexes [$sel get index]
> >>
> >>if you want them for a specific residue than you have to select
> >>something like this:
> >>
> >>set psiindexes [atomselect top "protein and resid 5 and name N CA"]
> >>set phiindexes [atomselect top "protein and resid 5 and name CA C"]
> >>
> >>Note that vmd outputs the indexxes in ascending order, not in the order
> >>you typed the atom names. If you need to identify individual indexes
> >>with atom names you have to select single atoms.
> >>
> >>Regards,
> >> Jan
> >>
> >>Paul R Brenner wrote:
> >>
> >>>Is it possible to use VMD to output index information for the dihedrals of
> >>
> >>a
> >>
> >>>molecule. I see that I can use the get function within atomselect to get
> >>
> >>the
> >>
> >>>angle values (phi,psi) but what I am looking for is an index.
> >>>
> >>>(I then plan to align these indexes with those assigned by a molecular
> >>
> >>dynamics
> >>
> >>>program to selectively restrict their variation.)
> >>>
> >>>Thanks for any pointers,
> >>>
> >>>Paul R Brenner
> >>>Graduate Research Associate
> >>>Computer Science and Engineering
> >>>The University of Notre Dame
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>
> >
> >
> >
> >
>
>