AMBER Archive (2002)

Subject: dummy parameters

From: Pierre Mignon (pmignon_at_vub.ac.be)
Date: Wed Oct 16 2002 - 02:10:46 CDT

Hello,
I am trying to compute the free energy change within the mutation of a tyrosine to an alanine. But i have problems when i try to save the perturbed parameters. Also i don't understand at all dummy parameterization..
The first thing I did was to define atoms to be perturbed and dummy atoms. Then when i tried to save the amber perturbed parameters, i had a error message saying that the dummy atoms were not defined. So i defined them as follow :
MASS
DH 1.008
DC 12.01

NONB
  DH 0.000 0.00
  DC 0.000 0.00

Then i tried to save the amber perturbed parameters but i had a error message saying that bond, angle, torsion and improper parameters were not defined for the dummy atoms. When i look to the tutorial “Methane in water box “ i can see bond and angle parameters as follow :

BOND
DH-DC 331. 1.09

ANGLE
DH-DC-DH 35. 109.5

But in the same time AMBER tells to parameterize torsion and improper with null force constants.

***The first question i have is why force constants for improper needs to be zero and for bonds or angles it is not null ? How must i defined these parameters ?***

Then i defined torsion and improper parameter as follow :
DIHEDRAL
DH-DC-DC-DH 1 0.0 0.0 3.
DH-DC-DC-DC 1 0.0 0.0 3.
CT-CT-HC-DC 1 0.0 0.0 2.
.
.
IMPROPER
CT-DC-HC-DC 0.0 0. 2.
HC-DH-DC-DC 0.0 0. 2.
DC-DH-DC-DC 0.0 0. 2.
.
.

Here is my second question :

***When i try to save the amber perturbed parameters i have the message that i don’t understand :

***Proper torsion parameters missing ***
 atom names : CG-CD1-CE1-HE1
 atom types : CA-CA-CA-HA =pert=> HC-DC-DC-DH
Please add a dummy parameter of multiplicity 3
for the non-pert types to your parameter set.
 -e.g. CA-CA-CA-HA 1 0.0 0.0 3.
(This is because multiple torsional potentials may be apply to a
single torsion, and each is perturbed individually in gibbs).

So I don’t understand why amber is asking for the parameterization of such torsion since they are already present in the parmxx.dat file.

Any help would be very appreciate, since one week, i can’t go further.
Thank’s in adavance.
Pierre

Mignon Pierre
PhD student
Free University of Brussel (VUB)
Dienst Algemene Chemie (ALGC)
Pleinlaan, 2
1050 Brussels
Belgium
Tel + 32 2 629 35 16
Fax + 32 2 629 33 17
e-mail pmignon_at_vub.ac.be