Re: alchemify FEP

From: Wei Chen (gtg553p_at_mail.gatech.edu)
Date: Fri Feb 18 2005 - 12:54:44 CST

Hi, Jerome,

Thank you so much!

However, when I ran the calculation, I met some problems. At the first step,
it was said that velocities of some atoms were faster than limitation. Then
the calculation was terminated. I had to reduce time step to 0.1 fs. Then
it could work. But it is too long to get calculation done. Do you have any
ideas?

Wei Chen

Quoting Jérôme Hénin <jerome.henin_at_uhp-nancy.fr>:

> Hi Wei,
>
> Derivation of the basic FEP expression shows that the kinetic contribution
> cancels out.
>
> Bonded terms are another problem. The FEP code in NAMD uses the (pretty
> common) assumption that their contribution is negligible.
> In the dual topology approach, the only bonded terms that would contribute to
> the free energy are those involving perturbed and non-perturbed atoms. In
> some systems, there will be no such terms (namely, when perturbing an entire
> molecule).
> Although I am sure one can imagine particular cases when this approximation
> make some small difference, but since the good use of alchemical FEP involves
> taking differences between free energy variations inside a thermodynamic
> cycle, most of this contribution is expected to vanish altogether.
>
> A good reference on that would be:
> S. Boresch and M. Karplus.
> The role of bonded terms in free energy simulations. 1. theoretical analysis.
> J. Phys. Chem. A, 103:103-118, 1999.
>
> Cheers,
> Jerome
>
> Le jeudi 17 Février 2005 20:14, Wei Chen a écrit :
> > Hi,all,
> >
> > I am using alchemify FEP in NAMD to calculate free energy difference before
> > and after mutating a residue in a peptide chain. However, I found that only
> > VDW and electrostatic energy are included in the calculation of free
> > energy. Why we do not need care about kinetic energy, bond stretching, etc.
> > Who have any ideas?
> >
> > Regards,
> >
> > Wei Chen
> >
> > --
>
> --
> Jérôme Hénin
> Equipe de Dynamique des Assemblages Membranaires
> Université Henri Poincaré / CNRS
> Tel : (33) 3 83 68 43 95 Fax : (33) 3 83 68 43 71
> http://www.edam.uhp-nancy.fr/
>
>

--

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:18:36 CST