Re: help generating psf files

From: Anna Modzelewska (amodzelewska_at_iimcb.gov.pl)
Date: Fri Oct 07 2005 - 02:58:19 CDT

Richard,
as far as I know this pdb is a result of the structure generation with an older topology file (it could be toph19.inp). The ace and cbx in that topology file were residues, now they are patches. In addition that top file used only the polar hydrogens.
If you want to create your pdb and psf for deca-alanin with acetylated and N-methylamidated terminus using the topology of charm22, I know two ways to do it:

1) You can remove ace and cbx entries from this pdb and you have to modify one command:
segment hin { pdb alanin.pdb
first ace
last ct3 }
In this way psfgen will add missing atoms and will correctly transform the terminus of the peptide. That's what I would do.

2) If you absolutely want to keep terminus coordinates from this pdb, you can modify its 6 entries:
ATOM 1 CAY ACE 2 -2.184 0.591 0.910 1.00 7.00 MAIN
ATOM 2 CY ACE 2 -0.665 0.627 0.966 1.00 0.00 MAIN
ATOM 3 OY ACE 2 -0.069 1.213 1.868 1.00 0.00 MAIN
ATOM 64 NT CBX 11 8.610 8.962 9.714 1.00 0.00 MAIN
ATOM 65 H CBX 11 8.050 8.324 9.225 1.00 0.00 MAIN
ATOM 66 CAT CBX 11 9.223 8.571 11.014 1.00 0.00 MAIN
Then use the same modification for segment line as above and in the lines before type:
alias residue ace ala
alias residue cbx ala

I hope I answered your question.
Anna

----- Original Message -----
  From: Richard Wood
  To: Anna Modzelewska
  Sent: Thursday, October 06, 2005 4:48 PM
  Subject: Re: namd-l: help generating psf files

  I guess the question I now have is since the NAMD
  people used this pdb file to generate a psf file, how
  did they do it?

  Also, ACE should be COCH3, not COC.

  I'm very familiar with topology files, as I used
  CHARMm for two years, and will most likely be using it
  again soon.

  Richard

  --- Anna Modzelewska <amodzelewska_at_iimcb.gov.pl>
  wrote:

> Indeed, ACE means that the N-term is acetylated. CBX
> is also a modification of the protein. They are not
> classic residues that's why psfgen can not find them
> in the residues section of the topology file. If you
> remove these two residues from your pdb and you try
> to create the psf in the same the way as usually you
> should success in.
> You can find ACE at the end of the topology file in
> the section with "patches".
> I recommend you to read the C Appendix of Namd
> Tutorial, the construction of topology file is well
> explained there.
> If you still have problems with psfgen let me know.
>
> Anna
>
>
> ----- Original Message -----
> From: Richard Wood
> To: Anna Modzelewska
> Sent: Thursday, October 06, 2005 3:30 PM
> Subject: Re: namd-l: help generating psf files
>
>
> Well, the residue ACE is incorrect in the file.
> It consists of three atoms: CA, C and O.
> I assume that residue CBX (N, H, CA) is incorrect
> also.
>
> Richard
>
> --- Anna Modzelewska <amodzelewska_at_iimcb.gov.pl>
> wrote:
>
> > Now, I'm sure that your pdb file is problematic,
> > could you send it to me? Do you have some
> > non-standard residues that you want to keep (as
> ACE
> > for example or some phosphorylated residues) in
> the
> > structure?
> >
> > Anna
> > ----- Original Message -----
> > From: Richard Wood
> > To: Anna Modzelewska
> > Sent: Thursday, October 06, 2005 3:12 PM
> > Subject: Re: namd-l: help generating psf files
> >
> >
> > hi Anna,
> >
> > It looks like it might be something to do with
> the
> > residue names, as when I type in segment ala {
> pdb
> > alanin.pdb }, it gives messages "unknown
> residue
> > type
> > CBX" and "unknown residue type ACE". Please
> note
> > that
> > this pdb file is one generated by the people
> that
> > develop NAMD. If I use their alanine
> parameters
> > and
> > this pdb file to run dynamics, it works fine.
> I
> > tried
> > to generaste a psf of this file because I had
> > tried
> > others (my own pdb files) and I got similar
> errors
> > when running psfgen.
> >
> > Richard
> >
> > --- Anna Modzelewska
> <amodzelewska_at_iimcb.gov.pl>
> > wrote:
> >
> > > Hi Richard,
> > > I'm not an expert, but that's what I would
> try
> > to
> > > type:
> > >
> > > topology top_all22_prot.inp
> > > segment ala { pdb alanin.pdb }
> > > coordpdb alanin.pdb ala
> > > guesscoord
> > > writepdb alanin1.pdb
> > > writepsf alanin.psf
> > >
> > > Remeber that you should precise the paths
> for
> > files
> > > that you use (pdb and topology file).
> > > The comment "Created by CHARMM version 22 1"
> is
> > > normal so I think that something else is
> wrong.
> > > Probably there is a problem with your pdb
> file.
> > > Please try to generate a psf for alanin with
> the
> > > commands above and let me know if it works.
> What
> > > version of VMD or NAMD do you use?
> > >
> > > Anna
> > > ----- Original Message -----
> > > From: Richard Wood
> > > To: Anna Modzelewska
> > > Sent: Thursday, October 06, 2005 2:47 PM
> > > Subject: Re: namd-l: help generating psf
> files
> > >
> > >
> > > Hi Anna,
> > >
> > > Well, all I did was run psfgen
> interactively.
> > I
> > > didn't use a script.
> > >
> > > First, I typed in topology
> top_all22_prot.inp
> > >
> > > (Note: when I do this, the output on the
> > screen
> > > ends
> > > with "Created by CHARMM version 22 1", and
> if
> > I
> > > type
> > > toplogy top_all22_prot.inp again, the text
> on
> > the
> > > screen goes as far as the line "duplicate
> > residue
> > > key
> > > LIG3 will be ignored")
> > >
> > > Then, coordpdb alanin.pdb
> > >
> > > When I do this, it gives me the following
> for
> > each
> > > atom : "Warning: failed to set coordinate
> for
> > atom
> > > CA
> > > CBX12 MAIN" (this is the last line for the
> > last
> > > atom)
> > >
> > > Finally, I type in writepsf test.psf,
> which
> > gives
> > > the
> > > following output:
> > > "total of 0 atoms
> > > total of 0 bonds
> > > total of 0 angles
> > > total of 0 dihedrals
> > > total of 0 impropers
> > > Info: psf file comple."
> > > (which it's not)
> > >
> > > This is so simple it should work, but it
> > doesn't.
> > >
> > > Richard
> > >
> > >
> > > --- Anna Modzelewska
> > <amodzelewska_at_iimcb.gov.pl>
> > > wrote:
> > >
> > > > Richard,
> > > > I've just tried and didn't have any
> problems
> > > with
> > > > creation of psf using the
> top_all22_prot.inp
> > > file.
> > > >
> > > > Could you send the script that you use
> as
> > the
> > > input
> > > > for psfgen and the output with error?
> > > >
> > > > Anna
> > > > ----- Original Message -----
> > > > From: Richard Wood
> > > > To: namd-l_at_ks.uiuc.edu
> > > > Sent: Tuesday, October 04, 2005 6:37
> PM
>
  === message truncated ===

  __________________________________
  Yahoo! Mail - PC Magazine Editors' Choice 2005
  http://mail.yahoo.com

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:19:03 CST