VMD-L Mailing List

From: Joshua Adelman (jla65_at_pitt.edu)
Date: Sun Nov 15 2009 - 07:06:58 CST

Hi Sarah,

It sounds like the problem is that the molecules/fragments you are having issues with, do not have corresponding entries in the topology file. AutoPSF needs some sort of residue definition telling it how the various atoms are connected, their charges, internal geometry, etc. If your residue isn't in the topology file, AutoPSF won't know what to do with those atoms.

I recommend taking a look at the following tutorial:
http://www.ks.uiuc.edu/Training/Tutorials/science/topology/topology-html/

You may have to either write your own topology file, or edit an existing one to include the missing residue types.

Best of luck,
Josh

On Nov 15, 2009, at 1:50 AM, sarah k wrote:

> hi Josh,
>
> my PDB files are created by HyperChem and Argus lab. I've tried simple molecules like naphthalene and indole. the 3 atoms of indole are 2 C and 1 H. the residue name was set automatically DEF. changing this name was useless. yes, there wasn't anything approximately relevant to my cases.
>
> recently, i receive these errors too:
> atomselect:cannot parse selection text: resname DEF and chain?
> when i try choose type in edit atom properties: VDW well depth (epsilon) must be a floating point number!
> Syntax error in expression "E?-1": variable references require proceding$
>
> in my previous tries i haven't faced them.
> if the problem has root in AutoPSF, then what's the solution? i don't have any acceptable excuse for my professor!! thanks.
>
> enjoy life,
> Sarah
>