Selected excerpts from results of the following commands are listed below:
cd amber.git/dat/leap
grep -r -w CY * > /home/lachele/amber.git.grep-CY
In the cmd directory (ignoring "oldff"):
cmd/leaprc.GLYCAM_06: { "CY" "C" "sp3" }
cmd/leaprc.ff02pol.r1: { "CY" "C" "sp2" }
cmd/leaprc.ff02polEP.r1: { "CY" "C" "sp2" }
cmd/leaprc.ff03.r1: { "CY" "C" "sp2" }
cmd/leaprc.ff03ua: { "CY" "C" "sp2" }
cmd/leaprc.ff10: { "CY" "C" "sp2" }
cmd/leaprc.ff99SB: { "CY" "C" "sp2" }
cmd/leaprc.ff99bsc0: { "CY" "C" "sp2" }
cmd/leaprc.ffAM1: { "CY" "C" "sp2" }
cmd/leaprc.ffPM3: { "CY" "C" "sp2" }
cmd/leaprc.ff99SBildn: { "CY" "C" "sp2" }
cmd/leaprc.ff99SBnmr: { "CY" "C" "sp2" }
In the parm directory, it is defined in:
parm/GLYCAM_04l.dat:CY 12.01 sp3 C aliphatic - for sialic acid only!
parm/GLYCAM_06EPa.dat:C C* CA CB CC CN CM CK CQ CW CV CR CA CX CY CD
parm/GLYCAM_06g.dat:CY 12.01 sp3 C aliphatic - for sialic acid only!
parm/opls_parm.dat:CY 12.01
parm/parm10.dat:CY 12.01 0.360 nitrile C (Howard et al.JCC,16,243,1995)
parm/parm91.dat:CY 13.02 ?
parm/parm94.dat:C C* CA CB CC CN CM CK CQ CW CV CR CX CY CD
parm/parm96.dat:C C* CA CB CC CN CM CK CQ CW CV CR CA CX CY CD
parm/parm98.dat:C C* CA CB CC CN CM CK CQ CW CV CR CA CX CY CD
parm/parm99.dat:CY 12.01 0.360 nitrile C (Howard et al.JCC,16,243,1995)
parm/parm99EP.dat:CY 12.01 0.360 nitrile C (Howard et al.JCC,16,243,1995)
parm/parmAM1.dat:C C* CA CB CC CN CM CK CQ CW CV CR CA CX CY CD
parm/parmPM3.dat:C C* CA CB CC CN CM CK CQ CW CV CR CA CX CY CD
Most of those redefinitions aren't a problem. You wouldn't normally expect to use two different protein force fields at once (tho I suppose someone might if different molecule classes were better treated in each). But several, like those in parms 91, 94, 96 and 98 (and some others like GLYCAM06EP) seem pointless. In fact, for the GLYCAM06EP, it's probably a bad thing since it is being equated to atom type C, which is sp2 (and CY is sp3 for glycam).
I think, in a number of these cases, certainly for glycam06ep, when parameter sets are copied over and edited, people haven't thought to delete unused information. As we start to simulate larger systems, though, ff combinations are likely to happen more often, so removal of extra stuff could become more important.
:-) Lachele
Dr. B. Lachele Foley
Complex Carbohydrate Research Center
The University of Georgia
Athens, GA USA
lfoley.uga.edu
http://glycam.ccrc.uga.edu
________________________________________
From: Yong Duan [duan.ucdavis.edu]
Sent: Sunday, December 11, 2011 6:16 PM
To: AMBER Developers Mailing List
Subject: Re: [AMBER-Developers] [AMBER] ATTN GLYCAM users: CY atom type work around
A list of atom types is easy to generate. In any case, there was not much
coordination before. We definitely need to improve.
Did you notice CY being used anywhere else? I noticed it is in
opls_parm.dat (for heme).
If you ask me, I think, in general, we should try to avoid redefining
existing atom types. Or, we'd expect conflict somewhere.
--
Yong Duan, Ph.D, Professor
UC Davis Genome Center and
Department of Biomedical Engineering
University of California at Davis
Davis, CA 95616
530-754-7632
On 12/11/11 3:03 PM, "B. Lachele Foley" <lfoley.uga.edu> wrote:
>I would still argue for a list because it will help people who are trying
>to decide which force fields can be used together. Since the Leap's
>don't complain if something is multiply defined, it becomes easy for
>parameters to silently be mis-loaded. And, while you can say "the user
>needs to learn to go look in the files," it seems we could make that a
>little easier. For example, if one merely looks at the defs at the top
>of parm94.dat, you would think CY isn't being used. And, it isn't --
>except for a van der Waals-type equivalence near the bottom.
>
>Perhaps CY wasn't a great choice, but it seemed as if parm91 was merely
>claiming it for some future use (as carbon-13, apparently), and its use
>had been mostly dropped by 94. In 94, its remaining as a vdW equivalent
>was probably just an oversight. So, it wasn't being used anymore, as far
>as we knew.
>
>What I'm saying is that if an atom type is used in something so old it's
>been hidden down in "oldff", should we still worry about reusing it? In
>04, the CY type hadn't been written significantly into a "parm" force
>field in 13 years. Maybe we shouldn't ever reuse, but there are only so
>many 2-letter combinations.
>
>We can't just load glycam as a frcmod afterwards. The trouble is that
>the definitions are too different. Our CY is sp3, and others are sp2.
>So, we have to load our atom types, then their atom types, then their
>parameters, then our parameters.
>
>:-) Lachele
>
>Dr. B. Lachele Foley
>Complex Carbohydrate Research Center
>The University of Georgia
>Athens, GA USA
>lfoley.uga.edu
>http://glycam.ccrc.uga.edu
>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Sun Dec 11 2011 - 16:00:03 PST
