On Fri, Aug 16, 2013, Ross Walker wrote:
>
> My opinion is that the first thing that needs to be done is to decide
> which of the polarization models is the path to follow
I think this has in large part been already decided: we have in ff12pol a
"simple" polarization model (point dipoles with Thole screening) for which
a very large effort has been put into parameterization (see all the work
from Yong, Junmei, Ray, Piotr and co-workers). On the other end, we have
(thanks to Bob Duke, Tom Darden, Celeste Sagui, Pengyu Ren and others)
perhaps the fastest Amoeba code, which is in turn a good starting point
for other "complex" ("advanced?") ideas such as SIBFA, GEM, etc.
Efforts that we can make to build on these signficiant existing efforts
will have immedate impact on simulations, and will surely provide us with
some experience that will serve us well when we get to the longer-range
ideas that Ross proposes (simple, local, easy to paramaterize, accurate,
works on 100,000 threads....).
>
> So I think we need to take a hard look at all the available polarizable
> models, how they work, and then rank them in terms of a series of criteria
Indeed. But while we are taking this hard look, I'm still hoping to find
people to help in making ff12pol and amoeba faster, more general, and
easier to use (including making some new tutorials).
...thx...dac
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Aug 16 2013 - 18:30:04 PDT
