Re: ABF force distribution

From: Kasra Fattah (kasra.fattah_at_gmail.com)
Date: Sat Mar 01 2014 - 18:20:11 CST

Mehran thanks for your help.

Jerome, actually what I did at the end was to use trajectory file binning
it and get the distribution of fs column (system forces). Basically the
running average of the fs at each cell along the reaction coordinate should
give the negative of fa (applied force) up to that point in time, is that
right?
The problem is that I was writing trajectory every 500 timesteps, so using
fs from the trajectory and average the values of fs in a cell is giving me
values which are different from the ones reported in *.grad file, either
I'm not doing the right job and have misunderstood the interpretation of
these values or this is because of working with forces at every 500
timesteps?!!
I would appreciate it if you comment on my analysis that is:
Should averaging the fs in trajectory file at each cell gives me exactly
the same value as what is reported in *.grad (negative of that basically)
if I use a trajectory that is written every timestpes?

Cheers,
Kasra.

On Thu, Feb 27, 2014 at 6:59 AM, Jrme Hnin <jerome.henin_at_ibpc.fr> wrote:

> Hi guys,
>
> the idea is that people who are interested in force distributions are also
> interested in force trajectories, and will save those. Then, it is fairly
> easy to extract the distributions from the time trajectories by binning.
> That's the reason why I didn't implement the force distribution calculation
> in the current ABF code.
>
> If however people feel strongly about this, feel free to present your
> arguments, and I could definitely be convinced to implement it.
>
> Best,
> Jerome
>
> ----- Original Message -----
> > Hi Kasra,
> >
> > I had a sample problem, it good to see someone did a little digging
> before
> > asking question. probably you find NAMD 2.6 user guid which talked about
> > 'distFile command'.
> > look at followig link for more info
> > http://www.ks.uiuc.edu/Research/namd/2.6/ug/node35.html
> >
> > fortunately, it works since it just need the abf TCL script which has
> been
> > already put in the newer version too. you can find it in
> >
> > $your NAMD folder/lib/abf
> >
> > However, it will give you the force distribution with fixed bin, 1
> > kcal/molA.
> >
> > -mehran
> >
> >
> >
> > On Thu, Feb 13, 2014 at 3:56 PM, Kasra Fattah <kasra.fattah_at_gmail.com
> >wrote:
> >
> > > Dear All,
> > >
> > > I'm trying to get the force distribution at each window for the ABF
> > > simulation that I'm doing. As I was digging through the list I found
> > > "distFile" command which seems is not available anymore in the recent
> > > versions of NAMD. Would you let me know how I can get the force
> > > distribution at each cell along the reaction coordinate?
> > >
> > > Cheers,
> > > Kasra.
> > >
> >
>

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:20:34 CST