VMD-L Mailing List

From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Tue Apr 21 2009 - 11:23:10 CDT

On Tue, 2009-04-21 at 12:29 -0300, Daniel Gaston wrote:
> Thank you for the reply,

daniel,

you can retrieve and set the secondary structure codes
at you liking via the script interface and atom selections.

set fixme [atomselect top "residue XX to YY and name CA"]
$fixme set structure L

those will override the output of stride.

cheers,
  axel.

>
> In regards to option b, I should have mentioned that I have already
> subjected these structures to minimzation and and MD tun (1.5 ns or so
> worth). The beta sheet predicted structure in that case has been reduced
> but not eliminated entirely, at least when done in a water sphere. I
> believe when I minimize and do the MD in a vacuum (done initially by
> mistake) that it was resolved.
>
> Option A does work. Ideally I would like to break and resolve that loop
> altogether. Oddly enough this is a case where I submit the same
> alignment to Swiss-Model using the same template structure (a homodimer)
> and only change which Chain is used as the template (A versus B). The A
> chain model works perfectly, it is only on the B chain that this loop
> appears to be modeled incorrectly.
>
> Thanks for the help. Any further suggestions/comments from users would
> be appreciated.
>
> Dan
>
>
>
> L. Michel Espinoza-Fonseca wrote:
> > Hi Daniel,
> >
> > There are two options I can think of:
> >
> > (a) You can use the 'new ribbons' rendering instead of the 'new
> > cartoon' one. You can change the thickness of the ribbons to make it
> > look like a cartoon but without worrying about the sheet depiction.
> >
> > (b) This option is more difficult, but better in terms of cleansing of
> > incorrect geometries/contacts. Probably what you could do is to relax
> > your protein using MD and minimize the final coordinates. This should
> > give you an improved geometry. Remember that homology modeling often
> > produces 'dirty' geometries. In this case STRIDE is just processing
> > what you're giving to it, whether is correct or not.
> >
> > Cheers,
> > Michel
> >
> > On Tue, Apr 21, 2009 at 4:25 PM, Daniel Gaston <daniel.gaston_at_dal.ca> wrote:
> >
> >> Hi all,
> >>
> >> I am wondering if there is a method to override STRIDE's secondary structure
> >> prediction (and thus how VMD displays it in New Cartoon format) for only a
> >> specified portion of a structure file? I have a Homology Model from
> >> Swissprot where one loop region in only one chain of a homodimer is show as
> >> as anti-parallel beta-sheet instead of the loop region that it should be.
> >> Because only this one portion is incorrect (and because Swissprot homology
> >> models contain only coordinates and not a secondary structure section) I
> >> would like to override the representation only for that one area of the
> >> structure without needing to create a secondary structure entry in the PDB
> >> for the entire protein. Is this possible?
> >>
> >> Thank you,
> >>
> >> Dan Gaston
> >>
> >> --
> >> Daniel Gaston
> >> PhD Candidate, Dr. Andrew Roger Lab
> >> Department of Biochemistry and Molecular Biology
> >> Dalhousie University, Nova Scotia, Canada
> >>
> >>
> >>
> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey_at_cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.