On Tue, Mar 16, 2010 at 11:57 PM, Bill Ross <ross.cgl.ucsf.edu> wrote:
>> This will override the wild-card definitions in the parm file. This
>> is because this parameter was defined last. If, for instance, you did
>> not use a -3 for the first line, then the second line would overwrite
>> the first (because a positive value in the last column tells leap to
>> terminate the series). However, for those dihedrals that the
>> wild-cards match that your definition above does not, those will still
>> take the values defined by the wild-card.
>
> The handling of multiplicities complicates this picture if memory
> of leap's algorithms is accurate. Leap's assignment of dihedrals is
> one of the most complicated bits of code that I have worked on.
Agreed, which is why I drew on the ff99SB example (and checked to make
sure I was not mistaken). To illustrate, C -N -CT-C is defined in
both parm99.dat and frcmod.ff99SB. Both of these files are loaded in
leaprc.ff99SB with no further adjustments made. (from leaprc.ff99SB:
#
# Load the main parameter set.
#
parm99 = loadamberparams parm99.dat
frcmod99SB = loadamberparams frcmod.ff99SB
#
)
frcmod.ff99SB simply fixes 4 torsion angles (of which the one listed
above, at the very least, appears exactly the same way in parm99.dat).
I realize dihedrals are far more complex because you have to allow
for multiple amplitudes, phases, and prevent overcounting. However,
as the usage in ff99SB indicates, these values are overwritten if
defined multiple times.
It was easier to look for an example that suggested the experts
thought it acted this way (by making ff99SB the way they did) rather
than dig through the source code :). (However, that dropped my
certainty to 99%, which is possibly naive overconfidence)
All the best,
Jason
>
> Bill
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
---------------------------------------
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Mar 16 2010 - 21:30:03 PDT
