Re: [AMBER] PM6-DH+ simulation problem

From: Andreas Goetz <agoetz.sdsc.edu>
Date: Thu, 11 Dec 2014 15:51:14 -0800

Hi Marcos,

Terms in this hydrogen bond correction depend on the geometry and connectivity of donor/acceptor groups. Your simulation encountered a geometry in which the computed connectivity (depends on the QM extracted bond order) is lower or higher than has been parameterized. Maybe try doing a classical geometry optimization first to make sure you have a reasonable starting geometry. Or a QM/MM geometry optimization with hydrogen bond correction. Note that it is possible to encounter this problem later on during MD.

All the best,
Andy

--
Dr. Andreas W. Goetz
Assistant Project Scientist
San Diego Supercomputer Center
Tel  : +1-858-822-4771
Email: agoetz.sdsc.edu
Web  : www.awgoetz.de
On Dec 10, 2014, at 9:05 AM, Marcos Serrou do Amaral wrote:
> Hi guys,
> 
> I am trying to use PM6-DH+ (in AMBER12) for a QM/MM simulation.
> I have two similar systems. In the first one, the solvent is methanol and
> the simulation works fine!
> 
> In another one, the solvent is ethanol and I have received the follow
> message in the first step for minimization:
> 
> "SANDER BOMB in subroutine calc_h_corr2 (dh_correction_module)
> Unusual bond definitions for H-bond correction.
> Bye bye."
> 
> and the simulation stops.
> 
> The solutes are very similar.
> Something like this:
> solute1 = solute + CH3 and solute2 = solute + CH2 + CH3
> 
> 
> Do you know where could be the problem ?
> 
> Cheers
> ---
> Marcos
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

Received on Thu Dec 11 2014 - 16:00:02 PST