Does your QM region span multiple (unbonded) molecules? If so, iwrap=1 can
cause one molecule to wrap around the box while leaving the other in place,
which can lead to a QM crash. (The MM routines consider nearest neighbors,
but I've seen the QM routines not do so, or at least terminate a simulation
before it even tries). We tend to recenter our restart files around one of
the QM molecules to avoid this problem.
Cheers,
Matt Zwier
On Thu, Mar 14, 2013 at 2:14 PM, R. Xiong <rchxiong.gmail.com> wrote:
> Yes, that was the message. So I just got lucky by using iwrap=0? No way to
> solve this problem? Thanks.
>
>
> On Thu, Mar 14, 2013 at 1:29 PM, Gustavo Seabra <gustavo.seabra.gmail.com
> >wrote:
>
> > If I had to guess, I'd guess that you have a QM region such that
> >
> > (QM size + cutoff) > (0.5 * box side),
> >
> > such that in periodic boundary calculations the QM region can see its
> > image, which is not allowed in Amber. However, this should be independent
> > of iwrap: it would appear anytime you try o use periodic boundary
> > conditions.
> >
> > Could you reproduce the actual message here? That would make the guessing
> > lot easier.
> >
> >
> > Gustavo Seabra
> > Professor Adjunto
> > Departamento de Química Fundamental
> > Universidade Federal de Pernambuco
> > Fone: +55-81-2126-7450
> >
> >
> > On Thu, Mar 14, 2013 at 11:42 AM, R. Xiong <rchxiong.gmail.com> wrote:
> >
> > > I am running QM/MM simulations using Amber with iwrap=1. It was running
> > > successfully, but when I restart the program, the program crashes
> because
> > > of QM/MM region is too large. I believe this should be caused by iwrap
> > > option. If iwrap=0, this problem will disappear, but my question is
> that
> > if
> > > there is any way to deal with the problem when still using iwrap=1.
> > > _______________________________________________
> > > 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
> >
> _______________________________________________
> 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 Mar 14 2013 - 12:00:04 PDT
