From: Jérôme Hénin (jerome.henin_at_uhp-nancy.fr)
Date: Tue Feb 28 2006 - 12:50:00 CST
> Don't you worry that fixing the ions might give you the wrong
> free energy value? In the simplest case, what if the dynamics
> of the ion were important for its interaction with the solvent?
I would say the dynamics of the ion is probably important for the *dynamics*
of its interaction with the solvent, but fixing it should not change the
statistical properties of the trajectory.
> I have done a series of FEP calculations with Na to K in water
> and I get very different shapes for free energy versus lambda
> if i restrain the ions to a fixed position space with different size
> spring constants. Most notably the spread in the calculated
> energies is very small with a strong spring, but it is more varied
> with a weak spring.
This sounds like a convergence issue, rather than a difference in the
calculated quantities. On the other hand, convergence does not account for
the different shapes of the free energy vs. lambda curve - unless each
lambda-window is short enough that its outcome is influenced by the
relaxation properties of ion-solvent interactions.
-- Jérôme Hénin Equipe de Dynamique des Assemblages Membranaires Université Henri Poincaré / CNRS Tel : (33) 3 83 68 43 91 Fax : (33) 3 83 68 43 87 http://www.edam.uhp-nancy.fr/
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:41:40 CST