VMD-L Mailing List

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Sat Nov 21 2009 - 08:59:35 CST

Hi Sarah,
all of the names of atoms in your input pdb *must* match the names of
the atoms in the topology file; otherwise psfgen has no idea how to map
the atoms in the pdb to those in the topology. Because you don't have
this naming scheme, most of the atoms cannot have coordinates assigned.
You should look at the "Parameterizing a novel residue" and "Topology
file" tutorials at http://www.ks.uiuc.edu/Training/Tutorials/ to get an
idea of how to do what you're trying here.
Best,
Peter

sarah k wrote:
>
> Hello all,
>
> I'm working on an organic compound. Its topology file wasn't
> available. So I used QMtool and defined atom properties, autogenerated
> all coordinates and wrote the Gaussian file. Then I saved the new pdb
> file as qm.pdb. After that I used paratool. My parent molecule was my
> first .*pdb file and I chose my created qm.pdb file as my base
> molecule. Then again I edited atom properties just like before and
> then I autogenerated coordinates. Then I wrote the topology and
> parameter files.
>
> Before doing such things, autopsf builder would generate an error
> saying molecule destroyed by fatal error. But now it does create the
> _autopsf.psf and _autopsf .pdb files successfully. Actually my origin
> file is a complex containing an organic compound plus a protein. The
> point is that when the psf file is written successfully (as the
> program claims!) we see a perfect molecule in the window. When I open
> the complex file or the autopsf.pdb file of the compound, I see a
> destroyed molecule with just several atoms.
>
> I've enclosed the *.pdb, *qm.pdb, *autopsf.pdb files of the compound
> and the topology file created by paratool. (I should add that when I
> use the complex, I add the topology file of the compound to the origin
> topology file, top-protein….inp, and save the file in wordpad by the
> format of *.txt. and then I use just the new txt file when running psf
> builder.)
>
> I have to solve the problem even if I there be no way but programming
> a new NAMD! Thanks in advance for your help and attention.
>
> Enjoy life,
>
> Sarah
>