On Fri, Jan 8, 2016 at 9:00 AM, InSuk Joung <i.joung.gmail.com> wrote:
> Unfortunately, changeLJSingleType() does not work.
>
> In omm_nonbonded_force() in structure.py, it seems that sigma and epsilon
> are actually used rathter than values in low-level parm arrays.
>
> So, I tried
>
> mineps = 0.01
> minsig = 0.1
> for a in prmtop.atoms:
> if a.epsilon_14 == 0.0:
> a.epsilon_14 = mineps
> print 'EPS14', a.epsilon_14
> if a.sigma_14 == 0.0:
> a.sigma_14 = minsig
> print 'SIG14', a.sigma_14
> if a.epsilon == 0.0:
> a.epsilon = mineps
> print 'EPS', a.epsilon
> if a.sigma == 0.0:
> a.sigma = minsig
> print 'SIG', a.sigma
>
> For some reason, a new value cannot be assigned to epsilon_14. The print
> line tells that a.epsilon_14 is not changed. (a.epsilon and a.sigma are
> replaced) I thought it is the source of the problem.
>
> Then I found this:
> In line 678, parmed/topologyobjects.py
> self._rmin14 = value
> should be replaced with
> self._epsilon14 = value
>
> Now, it works fine!
>
Oh man that's egregious! Thanks for catching this! I'll get this fixed.
It doesn't require a bugfix here, though, since you are clearly using the
Github version of the code :). I also need to get changeLJSingleType
working more generally for things besides prmtop modification.
Thanks again,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jan 08 2016 - 06:30:07 PST
