Re: Fix atoms-Colvars

From: Jérôme Hénin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Fri Jan 07 2011 - 13:04:10 CST

Hi Alex,

I don't understand what you are trying to do with colvars 3 to 6. If
you want to restrain the absolute positions of some groups, there are
two choices:
1) NAMD's per-atom "constraints"
2) a distance colvar defined between the group you want to restrain
and a "dummyAtom" at the target position. Then add a harmonic bias on
that (or just use the upperWall parameter).

Cheers,
Jerome

On 6 January 2011 22:46, Alex Liu <liu.alex77_at_yahoo.com> wrote:
> Dear NAMD users,
> I want to move an ion along z axis using dynamic z axis as explained in
> component "distanceZ" in NAMD 2.7b3 and get the PMF profile using ABF.
> I also want to confine the ion movement in XY plane so, I am using component
> "distanceXY" and for both components I am defining "main", "ref" and "ref2".
> In addition to above, I want to fix both ends of the dynamic z axis (using a
> force constant) by fixing the atoms I am using to define this axis (the
> atom numbers I have already used to define "ref" and "ref2"), but I am not
> sure whether I am doing it correct or not. Here is what I do:
> I define 6 colvars:
> Clovars1 and 2 are for the ion movement:
> Colvar 1. distanceZ to define the dynamic Z axis for the ion pathway (main,
> ref, ref2). Here main is the atom number of ion.
> Colvar 2. distanceXY to confine the ion movement in XY plane (main, ref,
> ref2). Here main is the atom number of ion above.
> Colvars 3 and 4 are for fixing the ref atoms:
> Colvar3. distanceZ (main, ref, ref2).  Here main is the same as ref in terms
> of the atom selection so, one end of the dynamic axis is fixed in Z
> direction.
> Colvar4. distanceXY (main, ref, ref2).  Here main is the same as ref in
> terms of the atom selection so, one end of the dynamic axis is fixed in XY
> plane.
> Colvars 5 and 6 are for fixing the ref atoms:
> Colvar5. distanceZ (main, ref, ref2).  Here main is the same as ref2 in
> terms of the atom selection so the other end of dynamic axis is fixed in Z
> direction.
> Colvar6. distanceXY (main, ref, ref2).  Here main is the same as ref2 in
> terms of the atom selection so the other end of the dynamic axis is fixed in
> XY plane.
> Then to get the PMF profile for ion movement (1-D) I just need to define
> Colvar 1 in ABF part. Can anyone please tell me if this approach is correct?
> In NAMD 2.6, I did this by using something like below:
>
> constraints on
> consRef restrain_ion.pdb
> consKFile restrain_ion.pdb
> consKCol o
>
> selectConstraints on
> selectConstrX on
> selectConstrY on
>
> selectConstrZ on
>
> So, I was thinking maybe instead of defining Colvars 3-6 in colvars
> configuration file to fix the atoms at both ends of the dynamic axis, I can
> simply use commands above in namd configuration file and just define Colvars
> 1 and 2 for the ion movement in colvars configuartion file. Is this also
> true?
>
> Or does anyone know another method I can use to fix these atoms?
>
> Thanks a lot for your time,
>
> Alex
>

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:19:42 CST