From: Jerome Henin (jhenin_at_cmm.chem.upenn.edu)
Date: Tue Nov 13 2007 - 11:37:27 CST
Vlad,
I tried to answer you questions below. I should add that I am still
wondering whether your problems might be solved by using a more
appropriate variable as a reaction coordinate...
On Nov 13, 2007 10:39 AM, Vlad Cojocaru
<Vlad.Cojocaru_at_eml-r.villa-bosch.de> wrote:
>
> Dear Jerome, NAMD users,
>
> I thought that the best option would be to introduce the component of the
> bias orthogonal to the direction of the reaction coordinate via an external
> tcl script while the ABF bias will be added on the r.c direction. However,
> in the ABF manual, its written that one should not call TclForces in the
> NAMD config file while running ABF. Could you give me a short hint about
> how should I invoke such tcl script while running ABF?
You can't define a separate script, so my suggestion is to add a
custom restraint to the existing ones in the ABF code. There are two
routines involved: restraints_init and restraints, towards the end of
the file abf_script.tcl. You need to add a section in each of these.
Hopefully the existing restraints will provide a clear template.
> And a second short question: I would like to add this additional bias only
> at the time steps when the ABF bias is apllied.
I don't know why you would do such a thing...
> How could I check if in the
> next time step an ABF bias will be applied (since "fullSamples" > 0, the ABF
> bias will not be applied continuously)?
One thing you could do is to add a parameter to the "restraints"
procedure, saying whether or not an ABF bias is being applied.
Jerome
> Jerome Henin wrote:
> Dear Vlad,
> I am not sure how to translate your idea in "theoretically correct" terms.
> My feeling is that the biasing force component orthogonal to v is
> arbitrary and not accounted for in the equations you will explicitly
> or implicitly be using.
> Have you tried simply applying the bias along v? If yes, and the
> behavior is not what you expect, then what you want to do is really
> add an external bias, *independent from ABF* to contain the system
> within a region that you consider relevant.
> I would avoid inserting that extra bias into the ABF one for clarity
> and to prevent breaking things in a subtle and undetectable way.
> Nothing prevents you from adding that bias using the existing
> restraints or by adding a custom one to that section of the code.
> Best,
> Jerome
>
> On Nov 8, 2007 9:28 AM, Vlad Cojocaru
> <Vlad.Cojocaru_at_eml-r.villa-bosch.de> wrote:
>
>
> Dear Jerome, Chris, NAMD users,
>
> Is it possible in ABF to apply the biasing force on a different
> direction than that of the reaction coordinate?
>
> I was thinking to retrieve the force $force (along the reaction
> coordinate (r.c.) in the "ABForce" procedure, then in ABFapply calculate
> a new force $new_force as $force/cos(alpha) (alpha = the angle between
> the direction of the r..c. and the direction of the applied force), and
> then apply this force on the new direction (Fvec = [vecscale $new_force
> $e_dir], where $e_dir = the unit vector of this direction).... In this
> way the force acting on the r.c direction will still be $force which
> will be used for the estimation of the free energy along the r.c. ...
>
> Such an application would make sense if I want to define the reaction
> coordinate as the projection of one vector v1 on a defined vector v in a
> protein and then move the head of v1 towards the head of v, rather than
> parallel to v ....
>
> Best wishes
> vlad
>
> --
> ----------------------------------------------------------------------------
> Dr. Vlad Cojocaru
>
> EML Research gGmbH
> Schloss-Wolfsbrunnenweg 33
> 69118 Heidelberg
>
> Tel: ++49-6221-533266
> Fax: ++49-6221-533298
>
> e-mail:Vlad.Cojocaru[at]eml-r.villa-bosch.de
>
> http://projects.villa-bosch.de/mcm/people/cojocaru/
>
> ----------------------------------------------------------------------------
> EML Research gGmbH
> Amtgericht Mannheim / HRB 337446
> Managing Partner: Dr. h.c. Klaus Tschira
> Scientific and Managing Director: Prof. Dr.-Ing. Andreas Reuter
> http://www.eml-r.org
> ----------------------------------------------------------------------------
>
>
>
>
>
>
> --
> ----------------------------------------------------------------------------
> Dr. Vlad Cojocaru
>
> EML Research gGmbH
> Schloss-Wolfsbrunnenweg 33
> 69118 Heidelberg
>
> Tel: ++49-6221-533266
> Fax: ++49-6221-533298
>
> e-mail:Vlad.Cojocaru[at]eml-r.villa-bosch.de
>
>
> http://projects.villa-bosch.de/mcm/people/cojocaru/
>
> ----------------------------------------------------------------------------
> EML Research gGmbH
> Amtgericht Mannheim / HRB 337446
> Managing Partner: Dr. h.c. Klaus Tschira
> Scientific and Managing Director: Prof. Dr.-Ing. Andreas Reuter
> http://www.eml-r.org
> ----------------------------------------------------------------------------
>
>
>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:31 CST