[gmx-users] Determining Force Constants for CG modelling

[Fabian Casteblanco]

Hello community,

I'm trying to create a topology for a molecule using the MARTINI force
field which is a coarse-grain (CG) forcefield.  I understand that to
optimize the bonded parameters, one needs to model the AA version and
extract the equilibrium angle and force constants.  As is said in the
MARTINI manual, an example is "the angle distribution function of a CG
triplet can be directly compared to the distribution function obtained
from the AA simulation of the centers of mass of the corresponding
atoms".

Once you find the pseudo-CG beads from the AA model, I see how you
would get the equilibrium parameters, but not the force constants.  Is
it simply taking the bonded potentials from the AA simulation and have
them matched to the CG bonded potentials by changing the force
constants so they have the same potential?  Is it necessary to use the
Boltzmann Inversion Approach?

Thank you for your help!


--
Best regards,

Fabian F. Casteblanco
Rutgers University --
Chemical Engineering