Subject: Re: [modeller_usage] short interatomic distances
From: Modeller Caretaker <>
Date: Tue, 29 May 2012 10:05:15 -0700
Cc: modeller_usage <>
On 05/26/2012 07:16 AM, M wrote:
I have a problem concerning the step of evaluating my model. I'm
using the whatif server
(http://swift.cmbi.ru.nl/servers/html/index.html). Most of the scores
are fine, but there are many abnormally short interatomic distances
according to whatif. The problem doesn't result from the alignment.
Is it possible that it is a result of different parametrization used
by modeller (or by the implemented CHARMM force field)?
The CHARMM force field in Modeller is fundamentally the same as in
CHARMM 22 itself, so this is unlikely to be the problem. The other terms
in the force field derive from the alignment, so if your model is poor
either the problem really is in the alignment or the optimizer failed to
satisfy all the restraints. The latter problem is generally solved by
building a larger number of models.