On Fri, Nov 1, 2013 at 11:56 AM, Jason Swails <jason.swails.gmail.com> wrote:
> Lachele put together a nice table that's in the
> doc/AtomTypesTableWorkspace.lyx file. It is a long list that appears to
> have taken a long time to put together. It includes most of the force
> fields (except ff13, it appears). It was added to the repo Feb. 2, 2012
> and last updated (by Lachele) Feb. 23, 2012. Those were 2 of the only 3
> commits (together with James' addition of ff12SB atom types).
>
> This was an excellent start, but unfortunately probably suggests to me
> that either the central database will not work, or the lyx table is the
> wrong medium to be using given its lack of attention and notoriety. I'd
Can't this doc just be converted to html and placed on the Amber website?
> This already exists to some extent in LEaP. Each atom knows about its
> hybridization, number of bonded partners, and atomic number. Hopefully
> this is enough to unambiguously assign GB parameters?
Yes, but then we're again locked into having LEaP determine what an
atom is from other parameters. If you assign an atom type an
unambiguous type from the get-go I think that's better as far as
future-proofing. These values then no longer need to be hard-coded in
LEaP which in my opinion is the way to go. If there is an unambiguous
atom type then there can be separate table files for each of the GB
radii set. So using my previous atoms it would look something like:
mbondi.dat
#UnambigiousType GBradius description
0 1.7 <some text, reference, etc>
1 1.3 <some text, reference, etc>
Using this format you could also have e.g. all the LCPO parameters in
one place, so they don't need to be hard-coded into every program
separately like we do now (which we've already seen is error-prone).
-Dan
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Nov 01 2013 - 12:00:03 PDT
