[MMTK] question on printing out all non-bonded interactions

Naomi Fox fox at cs.umass.edu
Sat Feb 4 19:54:38 UTC 2012


Konrad, thanks for your thorough explanation.  I think I'm going to stick
with the slow O(n^2) way of determining energies for all included nonbonded
pairs.  Using the NonbondedList seems a little tricky.  It involves reading
through the C code, and the script (copying from your two examples) keeps
getting seg faults.

I was wondering why when I sum the LJ-potentials over all atom pairs which
are non  excluded_pairs or one_four_pairs, I get a total of 561 kJ/mol.
 The LJ-potential for the system from universe.energyTerms() is 31 kJ/mol.
 This agrees with the LJ-potential I get when I sum the LJ-potentials
between each pair of residues.  Why the discrepancy in the energies?

Thanks!
-Naomi


On Thu, Feb 2, 2012 at 12:33 AM, Konrad Hinsen <
research at khinsen.fastmail.net> wrote:

> Naomi Fox writes:
>
>  > I would like to efficiently print out each non-bonded atom pair
> interaction
>  > and its associated LJ and coulombic potentials.
>  >
>  > I have used the example the MMTK file 'Examples/Miscellaneous/
>  > force_field_parameters.py' as a template.
>
> That's a good start, and in fact the only way to obtain the parameters
> at the Python level.
>
>  > How do I get all the atom pairs with a distance below some cutoff?
>  > Do the forcefields, for example Amber99, maintain some internal
>  > list of these?
>
> Not the force fields (they can't), but the energy evaluators do.  An
> energy evaluator is the result of applying a force field to a
> universe. Applying an energy evaluator to a configuration then yields
> the energy.
>
> The energy evaluators use a NonbondedList object to track pairs of
> atoms within a given cutoff as they move in a Molecular Dynamics
> simulation. NonbondedLists weren't meant to be used directly in user
> scripts, which explains that their interface is not the most
> convenient one, but it is certainly possible to do so and two examples
> have been discussed earlier in this forum:
>
>   http://starship.python.net/pipermail/mmtk/1999/000129.html
>
>   http://starship.python.net/pipermail/mmtk/2007/001306.html
>
> Konrad.
> --
> ---------------------------------------------------------------------
> Konrad Hinsen
> Centre de Biophysique Moléculaire, CNRS Orléans
> Synchrotron Soleil - Division Expériences
> Saint Aubin - BP 48
> 91192 Gif sur Yvette Cedex, France
> Tel. +33-1 69 35 97 15
> E-Mail: research AT khinsen DOT fastmail DOT net
> http://dirac.cnrs-orleans.fr/~hinsen/
> ---------------------------------------------------------------------
>
> _______________________________________________
> mmtk maillist  -  mmtk at starship.python.net
> http://starship.python.net/mailman/listinfo/mmtk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://starship.python.net/pipermail/mmtk/attachments/20120204/ab95fd76/attachment.html>


More information about the mmtk mailing list