From: Peter Freddolino (
Date: Mon Feb 15 2021 - 11:07:34 CST

Hi Manas,
I think it is really important for us to know what your end goal is for
this calculation. What energy are you really looking for? NAMDenergy is
giving you the actual nonbonded energy (Coulomb + LJ) of the interacting
groups in the force field of your calculation. If you're looking for
something like the contribution of a salt bridge to the stability of a
particular confirmation, or inter-molecular complex, you need to be getting
into the world of free energy calculations. This is beyond what namdenergy
provides. Just coming up with an ad hoc dielectric constant is unlikely to
help the situation.

On Mon, Feb 15, 2021 at 10:38 AM Manas Kohli <> wrote:

> Dear Daniel,
> Thank you for your email! I realised that the interaction energy is indeed
> what I was after and after some more reading I realised I needed to adjust
> the dielectric constant to get a more physiologically representative
> result. My salt-bridge exists at the extracellular face of the protein i.e.
> it faces water. Would you know of an appropriate dielectric constant to use
> for this calculation? Thank you in advance!
> Best wishes,
> Manas
> On Mon, Feb 15, 2021 at 1:23 AM Daniel Fellner <>
> wrote:
>> That seems about right, what's the issue?
>> *Daniel Fellner BSc(Hons)*
>> PhD Candidate
>> School of Chemical Sciences
>> University of Auckland
>> Ph +64211605326
>> On Mon, Feb 15, 2021 at 2:56 AM Manas Kohli <>
>> wrote:
>>> Hi all,
>>> I'm trying to calculate the free energy between two residues in a
>>> salt-bridge in my protein structure. I used the NAMD energy plugin and for
>>> input I specified protein and resid x and similarly protein and resid y for
>>> the two residues to select for. According to the NAMD energy documentation,
>>> the output should be in kcal/mol but I seem to be getting values in the
>>> range -100kcal/mol for an individual salt-bridge energy calculation. Am I
>>> doing something wrong in the calculation phase or specifying the wrong
>>> input? I know it may be difficult to troubleshoot the problem without
>>> seeing the structure but if anyone has a general idea, I'd appreciate it
>>> and can attach sample pdbs and similar information in subsequent emails if
>>> that helps.
>>> Thanks and Best Regards,
>>> Manas