AMBER Archive (2004)

Subject: AMBER: FEP - GIBBS of charged sidechains using IELPER

From: Annette Höglund (
Date: Sun Nov 28 2004 - 11:26:06 CST

Dear All,

There was some recent discussions about perturbing a charged sidechain
using the IELPER (electric decoupling) option.
One thing is still unclear to me:

Two legs are needed:
a: perturb only the charges of the sidachain atoms during A -> A'
b: perturb the atoms to dummy atoms (vdW (and what is rest of the charge
as well?)) during A' -> B

The question:
Is the A' the output from the first 1 -> 0 (lambda) perturb and then A'
(.crd) is used as input for a second 1 -> 0 perturbation?
Or does it mean that lambda is e.g. 0.5 at A', so that "two halves of
pert" are run (1-> 0.5 and then 0.5 -> 0)? In this case the lambda
should change from 1 to e.g. 0.5 in the first leg and then from 0.5 to 0
in the second? This means that there is still some charge to remove as
the atoms are changed to dummy in the second leg.

If the first is true, then it seems to me that the result is going to be
highly dependent on what A' is set to since removing the charge gives an
almost linear change of the F_energy (40-50 kcal), whereas the vdW
contribution seems to be very small (1-3 kcal).

Thanks in advance!


The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to