Another thing to keep in mind is that absolute energies are fairly
meaningless in most instances. What matters the most are energy
differences. In this case, the fact that all interactions between frozen
atoms are eliminated from the total potential doesn't actually matter when
it comes to comparing these differences.
If included, these interactions will not change at all between 2 different
conformations (as long as the belly definition is the same). Thus, if you
take a difference they would cancel out, anyway.
As for the 1-2' and related interactions, the 1-2 and 1-3 interactions are
omitted because the bond and angle terms take care of their interactions.
However, 1 and 2' are not actually bonded, so this is not the case
(therefore, they should still be included).
HTH,
Jason
On Fri, Jul 29, 2011 at 11:34 AM, case <case.biomaps.rutgers.edu> wrote:
> On Fri, Jul 29, 2011, Ignacio J. General wrote:
>
> > OK, that is clear now. But I have a last question. As far as I know,
> > in the Ewald equations it is not possible to ignore, say, 1-2
> > interactions without also ignoring 1-2', 1-2'', etc. interactions
> > (where the primes represent the copies of atom 2 in other cells). In
> > general, the effect of 1-2', 1-2'', ... is pretty small when compared
> > to 1-2, so it should be a good approximation. Is that what Amber does,
> > or is there some other way of ignoring 1-2 without ignoring 1-2', etc?
>
> Briefly (see the implementation papers for details): the reciprocal part of
> the PME calculation includes all interactions. After this, the reciprocal
> part of the 1-2 interaction is subtracted (in a sum over 1-2 interactions).
> Hence 1-2' interactions *are* a part of the total energy, but 1-2
> interactions are excluded.
>
> ....dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jul 30 2011 - 14:30:03 PDT
