[gmx-users] Free energy of turning on restraints

[Robert Johnson]

Hello everyone,
I'm computing the free energy of binding between a DNA base and a
carbon nanotube. I'm following the recipe given in Boresch et. al. J.
Phys. Chem. B. 107, 2003 where the ligand (the base in my case) is
restrained. The free energy of turning on the restraints is included
in the overall binding free energy. These restraints include 1
distance, 2 angle, and 3 dihedral restraints.

Unfortunately, I'm having problems including the correct restraints in
my topology file. First off, the form of the angle restraint in
Gromacs involves 4 (2 pairs) atoms. The restraint in the Boresch paper
assumes that the angle is between 3 atoms. Is there anyway to apply
the correct restraint in Gromacs?

Also, it doesn't seem like the [ angle_restraints ] or [
dihedral_restraints ] sections can be perturbed in a free energy
calculation. For example, when I include two sets of restraint
parameters (one for topology A and one for topology B) for, say,
dihedral restraints, I get the following error:
-------------------------------------------------------
Program grompp, VERSION 3.3.1
Source code file: toppush.c, line: 1180

Fatal error:
Incorrect number of parameters - found 9, expected 5 or 5.
-------------------------------------------------------

How do I perturb (i.e. turn on) these restraints over the course of a
simulation?

Thanks,
Bob Johnson