Hi All,
I've been adding LES trajectory splitting/averaging to cpptraj and have hit
a snag. Everything works fine if I'm using a "simple" LES topology
(something like copies of only one region). However, if I try to make a
more complex LES topology (varying #s of copies of several regions) I get a
topology that's not set up the way I had expected (details below). This is
probably just a case of me not understanding how an LES topology is set up,
so before I get into the details of my problem I was wondering are LES
topologies are described in-depth anywhere? I searched through the website
and GIT tree docs but was unable to find anything.
This is what I currently understand the LES-related flags in the topology
to mean:
LES_NTYP: Total number of LES types.
LES_TYPE: LES type for each atom
LES_FAC: LES scaling factor for interaction of typeA x typeB (essentially
an NTYP*NTYP matrix)
LES_CNUM: LES copy # for each atom
LES_ID: LES region ID, corresponding to the original 'spac' entries in
addles input
The stuff that's important for LES trajectory splitting is LES_CNUM; an
atom has CNUM of 0 if its in every copy, otherwise the CNUM is what copy
that atom belongs to. So what makes sense is that the number of 0 entries
plus the number of any individual copy should equal the total number of
atoms in the original (non-LES) topology. This seems to be the case for an
LES topology with input like this:
~ 3 copies of the fragment consisting of residues 191 to 215
spac numc=3 pick #mon 191 215 done
But not for an LES topology with input like this:
~ 3 copies of the fragment consisting of residues 191 to 215
spac numc=3 pick #mon 191 215 done
~ 4 copies of the fragment consisting of residues 234 to 243
spac numc=4 pick #mon 234 243 done
~ 2 copies of the fragment consisting of residues 42 to 55
spac numc=2 pick #mon 42 55 done
~ 3 copies of the fragment consisting of residues 55 to 60
spac numc=3 pick #mon 55 60 done
The latter input has CNUM entries that go up to 6; however the total for
each copy does not add up to the # of atoms in the original topology (in
this case 4049). This is how CNUM breaks down:
CNUM: #_of_atoms_with_that_cnum Total
0: 3232
1: 817 4049
2: 817 4049
3: 594 3826
4: 140 3372
5: 14 3246
6: 14 3246
The first two appear fine, but after that its strange. Note that the
ordering of the 'spac' entries doesn't appear to affect CNUM, but does
appear to affect the last number in POINTERS (NCOPIES I think?), which BTW
never equals 6.
Any ideas?
-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 Tue Dec 03 2013 - 16:00:02 PST
