AMBER Archive (2005)

Subject: Re: AMBER: Improper Torsional Parameters

From: David A. Case (case_at_scripps.edu)
Date: Wed Nov 02 2005 - 10:47:34 CST


On Fri, Oct 28, 2005, Ilyas Yildirim wrote:
>
> Does anyone of u know how the improper torsional parameters enter into the
> potential energy calculation?

The functional form is the same as for "proper torsions". LEaP identifies all
atoms that are three-coordinate, and looks to see if the parm file has an
improper defined for that situation. If so, it uses it; if not, it goes on.

> For instance lets say that we have the
> following parameters (extracted from parm99.dat file):
>
> CK-CB-N*-CT 1.0 180. 2.
>
> How are these parameters used in the energy calculation?
>
> In the parm99 file (for example), sometimes the wild-card is used, as it
> is done in the following example:
>
> X -X -C -O 10.5 180. 2. JCC,7,(1986),230
>
> The first example is the specific case while the second one is the generic
> case. In page # 257 of the AMBER Manual, it is talking about Dihedral
> Angle, and how these dihedral parameters are used to calculate the
> energies. I am wondering what the difference between generic and specific
> case is for improper torsions.

"specific" torsions are search for first, and used if a match is found. If
no match is found, then a search is made to see if a "generic" (aka wild-card)
torsion with match.

>
> As it is described in the manual, for the dihedrals, if PK=3.0 and
> IDIVF=9, then this is used to describe the generic case.

Note that, even the functional form for proper and improper torsions is the
same, they form distinct parameter sets. That is, for proper torsions, only
the proper torsion section in the parm.dat file is searched for a match.
Likewise, only torsions that are in the improper part of the parm.dat file
will be searched for matches for improper torsions.

If you are making frcmod files with improper torsions, be sure to apply
bugfix.53.

...good luck...dac

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu