RE: Free-energy perturbation in NAMD: Evaluation of electrostatic and van der Waals terms

From: Christian Jorgensen (
Date: Tue Nov 09 2010 - 09:42:22 CST

Hi again,
I have found a possible answer in Pohorille, A., Jarzynski, C., Chipot, C., Good Practices in Free-Energy Calculations, J. Phys. Chem. B, Vol. 114, No. 32, 10235-10253 2010 (

On page 10249:
Under certain circumstances, concomitant scaling of Coulomb and van der Waals interactions may lead to a situation in which the energy and forces are dominated by the former interactions before the repulsive component of the latter is large enough to avoid spurrious collisions of the constituent atoms. This shortcoming can be circumvented by decoupling electrostatic and van der Waals interactions. It is possible to scale Coulomb interactions with a sufficient delay to avoid scenarios in which Lennard-Jones repulsion is overwhelmed by electrostatic forces. Alternatively, the Coulomb potential can be modified in the spirit of eqs. 64 and 65 [for van der Waals interactions, edit] replacing the interatomic distance rij with a related soft-core term (αλ^2 + rij^2)^1/2

My question: Would this be the way that NAMD treats the relationship between the van der Waals and electrostatic terms?

Thanks so much
From: []
Sent: Tuesday, November 09, 2010 6:38 AM
To: Christian Jorgensen;
Subject: Re: namd-l: Free-energy perturbation in NAMD: Evaluation of electrostatic and van der Waals terms


Probably, it is not a good idea to suppose that every person on the list have the book that you are citing (yet they should, the are some very good
explanations in there). Anyways, the implementation of the soft core potentials for the alchemical calculations are here If i am understanding well what are asking, the files that you need is ComputeNonBondedFEP.C and ComputeNonBondedBase.C (but it was just a fast look).



----Mensaje original----
Fecha: 08-11-2010 19:23
Para: ""<>
Asunto: namd-l: Free-energy perturbation in NAMD: Evaluation of electrostatic and van der Waals terms

Hi all,
In the book Chipot and Pohorille "Free Energy Calculations", Springer, there is an equation (2.49) for the modification of the parametrization of the van der Waals term in the potential energy function that governs the interaction of an appearing or disappearing atom i with an unaltered one j.

My question: How is this equation coded in the src directory of NAMD?

I am only just starting my graduate studies and cannot read C / C++ so I do not entirely understand the files in the NAMD src



My question: I also want to know how NAMD codes the equivalent equation to (2.49) for the electrostatic potential?

Any input here would be much appreciated!

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:44 CST