Re: ABF parameters

From: Jerome Henin (
Date: Mon Oct 01 2007 - 10:53:41 CDT

Hi Vaithee,
Indeed, some keywords are missing from the manual. Thank you for
pointing this out.

writeFxiFreq allows one to print the instantaneous force Fxi to the
NAMD output with a given frequency.
df is the bin width when saving Fxi distributions (histograms) using
the distFile option.

The keyword temp is only used for reaction cordinates that require a
Jacobian correction to obtain uniform sampling. It is used to provide
the simulation temperature to the ABF code, which at the moment has no
way of querying it from NAMD. This should change in future versions.

Note that if ABF calculations are run with an incorrect value of temp
(in your case, 300 instead of 328), it does not introduce any error in
the results. The only consequence is that sampling along the RC is
slightly non-uniform - that is, for infinite simulations times. In
usual real-life cases, the effect will hardly be noticeable.


On 10/1/07, Subramanian Vaitheeswaran <> wrote:
> Dear NAMD developers,
> I am using the ABF module of NAMD to calculate potentials of mean force and interaction free energies. Can any one tell me what the parameters temp, writeFxiFreq and df mean? The User Guide does not say anything about these.
> I don't define these quantities in my input file, but the log file has:
> TCL: ABF> temp : 300.0 [default]
> TCL: ABF> writeFxiFreq : 0 [default]
> TCL: ABF> df : 1.0 [default]
> Seems like "temp" may be a temperature, but this is set to 328 and the log file does say:
> So, what is "temp" in the context of ABF?
> thanks,
> Vaithee

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:19 CST