From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Sat Nov 24 2012 - 06:17:07 CST
On Fri, Nov 23, 2012 at 9:29 PM, Steven W Rick <SRick_at_uno.edu> wrote:
> I am interested in using NAMD for simulating charge transfer models we have recently developed in our lab. To do this, I need to make
> some additions to the code. To do this I need to find the lines of code for all the electrostatic calculations, where
> the charge variable are kept, how the neighbor lists are kept, and how the forces are stored. If you could point me to the right places, that would be a big help.
steve,
i don't think that there is going to be much advice
outside of "you have to read the sources" and "use grep".
the structure of NAMD is somewhat different from other
MD codes, as it has to accommodate parallelization
using charm++ objects and thus one has not much
of an alternative and getting an overall appreciation
for the general flow of control.
is there a specific reason, why you have to be using NAMD?
there are other codes out there, which may be easier to
adapt, even if they won't scale/perform quite as well.
i also know of some cases/codes, where people have
already implemented charge equilibration schemes, so
you'd have a head start in implementing your method
by following a similar scheme.
> I would want to do this using standard PME.
>
> Specifically, to implement this I need to follow these steps.
> 1) Calculate the amount of charge transferred based on nearest neighbor distances (and so would need neighbor lists) and modify the charges of all atoms.
> 2) Calculated the electrostatic interactions, and also I need to calculated dE/dq (the derivative of the energy with respect to charge) for
> all charges.
> 3) Add the charge transfer contribution to forces and energy
have you considered how this would be doable with
parallelization via domain decomposition and particularly
how step 2 would be translated to using long-range
electrostatics?
cheers,
axel.
>
>
> Thanks!
>
>
> Steve Rick
> Professor of Chemistry
> University of New Orleans
>
>
-- Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0 International Centre for Theoretical Physics, Trieste. Italy.
This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:22:47 CST