VMD-L Mailing List

From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Wed May 20 2009 - 16:28:54 CDT

On Wed, 2009-05-20 at 15:49 -0400, Yi Shang wrote:
> Dear there,

dear miranda,

> I wanted to load pdb files generated by VMD into tleap (in AMBER
> package).
> But I notice that Hydrogen naming in pdb files generated by VMD, is
> different from those naming of pdb files generated by tleap.
> for VMD pdbs: 2HD, 3HD, 2HG, HD21, HD32.....
> for tleap pdbs: HD2, HD3, HG2, 1HD2, 2HD3......
> It's giving my a hard time when I use ptraj (another program in AMBER
> package)
> and now tleap.
>
> I wonder how these naming inconsistency happened. And hope I can get
> some suggestions here to get away with it.

as far as i know, this has nothing to do with VMD itself, but rather
the difference in naming conventions between charmm and amber. you can
easily write a small script that converts between one and the other.
it may be most convenient to put it into your .vmdrc

say you create a list of mappings:

set chm2ambmap = { {2HD HD2} {3HD HD3} {2HG HG2} {HD21 1HD2} {HD32
2HD3} ... }

global chm2ambmap

then you can do for the forward conversion:

proc chm2amb {molid} {
  global chm2ambmap

  set cnt 0
  foreach p $amb2chm {
    set sel($cnt) [atomselect $molid "name [lindex $p 0]"]
    incr cnt
  }

  set cnt 0
  foreach p $amb2chm {
    $sel($cnt) set name [lindex $p 1]
    $sel($cnt) delete
    incr cnt
  }
}

and for the reverse process

proc amb2chm {molid} {
  global chm2ambmap

  set cnt 0
  foreach p $amb2chm {
    set sel($cnt) [atomselect $molid "name [lindex $p 1]"]
    incr cnt
  }

  set cnt 0
  foreach p $amb2chm {
    $sel($cnt) set name [lindex $p 0]
    $sel($cnt) delete
    incr cnt
  }
}

and then just call the command

chm2amb top
or
chm2amb 1

whenever you want to convert a specific molecule's
naming conventions. there are similar ones for the
residue names, so you want to include this as well.

> Another question: the VMD pdbs are single chain molecule by default,
> how could I specify it to add 'TER' in between protein, ligand, or
> water, water?

right now you would have to write a small script that post
processes your .pdb file with, e.g., awk, perl, or python,
and split it apart based on some other information, e.g.
segment name or chain id.

making VMD insert these automatically would be conceivable
if there was a consensus as to what indicator would be
the correct one. there is, however, always the concern that
any choice would break other programs that have different
conventions.
i would not want to have TER records between individual waters,
though, as that would blow files up significantly.

cheers,
   axel.

>
> Thank you all!
>
>
>
> --
> Miranda 

-- 
=======================================================================
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.