Re: Vanishing bonds in small molecule during psf generation.

From: Peter Freddolino (
Date: Wed Jul 18 2007 - 14:19:26 CDT

Hi Christian,

Christian Blouin wrote:
> Hi,
> The problem at the time was indeed embarrassing oversight/typos. Thank
> you very much for the helping hand! I now have to deal with the fact
> that many of the bonds that I’ve specified aren’t parameterized in the
> par_all27* file. Also, for some reasons, atoms are added during psfgen
> which I suspect is an attempt to cap the polypeptide chain (even if
> there are none).
If one of your topology files defines automatic patching (which is
likely), you should explicitly turn it off for your cofactor segment:

segment X {
pdb cofactor.pdb
first none
last none

> Getting a cofactor up and running feels like a dark art! Is everyone
> modeling proteins with cofactors end-up with their own _/handcrafted/_
> top and par files?
Right now, pretty much. Paratool is an effort to change this in
vmd/namd, and I know amber has antechamber going along similar aims.

Also, it sounds like you got advice from several people who didn't post
on list. To everyone who helped out, it would be helpful in general for
replies to be posted on list unless they involve the exchanges of
nontrivially-sized attachments, so that everyone can benefit from the

> Christian

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:58 CST