Re: setting configuration file for free energy perturbation

From: Jérôme Hénin (
Date: Mon Jul 18 2005 - 12:18:39 CDT

Hi Deborah,

On Monday 18 July 2005 17:25, Deborah Jameson wrote:
> I wanted to do simulation for computing free energy perturbation in NAMD .
> I got the error message "Setting parameter fep file from script failed"
> after the 1000 step results for minimization .

Yes, you have to use two different config files to run the minimization and
the FEP calculation. You should also include a heating step, to leave your
minimized system enough time to reach the target temperature.

> fep on
> fepfile rand.fep
> fepCol B
> fepOutFile rand.fepout
> fepOutFreq 5
> fepEquilSteps 3200
> set step 0
> set dstep 0.025
> dlambda $dstep
> while {$step < 1} {
> firsttimestep 0
> lambda $step
> set step [expr $step+$dstep]
> lambda2 $step
> run 6400
> }

> Also, I want to ask about the difference between fepEquilSteps and run
> manual it is mentioned that in each window the esystem is
> equilibrated in 3200 steps before 6400 steps of data collection.if anybody
> help me on this problems and explain me more on the process of fep
> calculation,I will be very thankful.

Your script will result in 6400 steps per window, half of which will be
considered as equilibration and therefore not used for computing the free
energy change.

On "the process of FEP calculation", let's say that you want to compute the
free energy change for a (alchemical) transformation of your system. It is
much easier to compute this free energy change in pieces, by adding the
changes resulting from a number of small, successive transformations. That's
called staged FEP. To define stages, or windows, in the transformation, you
need to connect the initial and final states with intermediate states, and
the coupling parameter lambda is the simplest way of doing that. Each value
of lambda defines an intermediate state somewhere between the endpoints -
which are defined by lambda equal to 0 or 1. Thanks to the Tcl script in your
config file, NAMD will run a short dynamics at each of these lamba-points,
and compute the free energy change for the corresponding window.

The basic FEP theory is briefly presented in the NAMD manual, in our tutorial,
and in several textbooks on molecular dynamics. Now if you want more
information, you can ask in a more specific manner.


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 87
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
No, I am tired of advertisement.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:40:57 CST