From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Mon Jun 15 2009 - 09:45:56 CDT

How are you running psfgen? You should be able to make the entire
protein into one segment simply by only having one segment statement in
your psfgen script...
Best,
Peter

Víctor wrote:
> Thanks Axel.
>
> What I really want is to do this kind of "fixing" ( just getting some
> kind of more relaxed structure, because i'm working an iterative
> modelling algorithm).
> I've tried wit minimization fixing CA or N atoms with NAMD, but as they
> need also the psf files I need to process them. As the structure is
> somewhat chaotic in some points, psfgen tries to patch it where there is
> no need (for me) and it creates too many segments when I just want to
> have one (my single chain segment). Would it be possible to tell psfgen
> not to apply any patch? (i've read the u.g. and it seems it's
> impossible, isn't it?
>
> Regards,
>
> On Sun, Jun 14, 2009 at 11:58 PM, Axel Kohlmeyer
> <akohlmey_at_cmm.chem.upenn.edu <mailto:akohlmey_at_cmm.chem.upenn.edu>> wrote:
>
> On Sun, 2009-06-14 at 22:35 +0200, Víctor wrote:
> > Dear all,
>
> dear victor,
>
> > I have a self made protein structure with very bad positioned residues
> > and then the structure is about to be disconected in some parts. I've
> > noted that when i get the psf file with autopsf the on on screen
> > structure seems to be fixed, but the pdb file is still the same so i
> > think that the psf file is containing some kind of patches to fix the
> > structure...Can I combine this psf and pdb to get a pdb file with the
> > fixed structure??
>
> what you are seeing is that the .psf file contains the information about
> bonds explicitly, but the .pdb does not. if you provide a .psf file,
> then VMD uses the bond information from the file, with only a .pdb
> file, it has to guess. if your structure is not "good" it will guess
> wrong. the best way to handle this is to either load _both_ the
> .psf and the .pdb file (the .psf has no coordinate information) or
> save the structure and coordinates in a different format that stores
> both coodinates and topology information.
>
> > Or maybe you know another method for fixing structures?
>
> this is not really fixing the structure, it only changes the
> bonds according to your residues from the templates that psfgen
> knows. to "fix" the structure, you will have to do a geometry
> optimization or something alike.
>
> cheers,
> axel.
>
> > Thanks in advance,
> >
> > --
> >
> > Víctor Gil Sepúlveda
> > E. Informatica - FIB - UPC
>
> --
> =======================================================================
> Axel Kohlmeyer akohlmey_at_cmm.chem.upenn.edu
> <mailto: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.
>
>
>
>
> --
>
> Víctor Gil Sepúlveda
> E. Informatica - FIB - UPC