[gmx-developers] Writing the electric field (and its gradient) on particular atoms during an MD simulation

[Gerrit Groenhof]

Hi John,

Instead of modifying the code (which is more elegant of course), would 
it not be possible to do a rerun, with all bonded and Lennard-Jones 
interactions set to zero in the topology? I think that in that case the 
forces on the atoms, as printed in the trr file are equal to the 
electric field
(f(i)=q(i)*E(i), right?). Or do I overlook something here?

Depending on the accuracy you want, The gradient (Nable E) might be 
obtained by finite differences, once you have the electric field 
associated with each atom position, but I do not know how easy or cheap 
this is.



Gerrit




John Chodera wrote:
> Gentlemen,
>
> I am working with a collaborator (Thomas la Cour Jansen) who is
> developing methods for computing 2DIR spectra from molecular dynamics
> simulations, for purposes of comparison between simulations and
> experiment.  His methods employ a number of functions parameterized by
> (1) atomic coordinates, (2) the electric field impinging upon
> particular atoms, and (3) gradient of this field with respect to the
> position at which this field is measured.
>
> For these methods to work with a multitude of forcefields in a
> self-consistent way, it would be extremely useful if we could have
> gromacs write out the electric field on these atoms (and its gradient)
> every N steps.  To be consistent with the way in which energies and
> gradients for forcefields like AMBER are computed, the electric field
> computed at atom position i would exclude the contributions from atom
> i itself and the direct-space Coulomb interaction from atoms separated
> by one or two bonds, and the contribution from atoms separated by
> three bonds would be scaled (by half, in the case of AMBER).
>
> Most, if not all, of this information is presumably already computed
> in the force/gradient routines.  We are trying to assess how much
> effort it would be to extract this information and whether it is
> possible for a mere human like me to make these changes, or if I would
> need to appeal to one of Wise Developers to make such a modification.
>
> Many thanks in advance,
>
> - John
>
> --
> Dr. John D. Chodera <jchodera at gmail.com>      | Mobile    : 415.867.7384
> Postdoctoral researcher, Pande lab            | Lab phone : 650.723.1097
> Department of Chemistry, Stanford University  | Lab fax   : 650.724.4021
> http://www.dillgroup.ucsf.edu/~jchodera
> _______________________________________________
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-developers-request at gromacs.org.
>