Re: questions regarding output data of an ABF simulation

From: Jerome Henin (
Date: Mon Aug 07 2006 - 17:02:43 CDT

Dear Anahita,

I see from you config file that you have setup a broad reaction coordinate
interval with very fine bins (.01 Angstroms!), so unless you run an extremely
long simulation, it is not surprising that many of these bins end up being
The best way to see what happens is to use the output of "writeXiFreq" and
look at the xi(t) graph. The simplest way to do it is:
grep "Xi at timestep" abf-run.out | awk '{print $6, $8}' > xi_of_t.dat
and then open xi_of_t.dat with a plotting/spreadsheet program.
At the xi positions that the system actually visits, you should definitely
have *some* data.

In general, a bin width of .1 Angstrom is narrow enough to describe most PMFs

Please post back to this list any further problems you may encounter with ABF.


On Monday 07 August 2006 16:27, Anahita Tafvizi wrote:
> Dear fellow NAMDers,
> I'm seeking your help regarding a PMF calculation in NAMD using ABF.
> I am trying to run ABF for lacI protein sliding along the DNA. I define my
> abf1 and abf2 as the last pair of DNA and the protein respectively. the
> last pair of DNA is completely fixed (occupancy= 100.0), and the backbone
> of the rest of the DNA is also fixed but with a smaller occupancy (5.0).
> The protein is moving toward the last pair as I want, but I don't have any
> useful output data. I mean I can see the movie, but I my history file is
> just containing zeros.
> I've attached my namd file.
> Thanks a lot in advance for your replies.
> Best regards,
> Anahita

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:43:54 CST