[gmx-users] SMD : pulling both primes of DNA

[Thomas Schlesier]

The easiest solution would be using 'pull_geometry = distance' in all 
three dimensions. Then you can be sure that both groups are pulled 
together. Small remark: One group would be fixed for the pulling, and 
the second group gets pulled towards the first group. So if you want to 
have both groups really close, this setup would be the easiest solution.

If you really to pull both groups to a point in the middle between them, 
i think you would need 'pull_geometry = position':
Use some part of the DNA as the reference group and put the origin of 
the pulling  (via 'pull_init') to the middle point. Then make two 
pull-groups + 'pull_vec1' or 'pull_vec2', from each DNA part, which you 
want to pull to the middle point.
In principle this setup should work, but you can run into problems if 
the structure of the DNA changes and the middle point, to where you pull 
both groups, moves somewhat around.
hard to explain...

I would stick to the first approach...
What pull-mode did you use? You could use 'pull = constraint', measure 
the distance between both groups and then pull with a given pulling 
velocity so long till both groups should meet. Since you use a 
constraint instead of a restraining potential, the groups should be 
definitly pulled together (or some other parameters in the .mdp file are 
wrong).

Greetings
Thomas






Am 27.03.2013 20:16, schrieb gmx-users-request at gromacs.org:
> Hello All,
>
> My simulation system is composed of DNA and want to pull both of the primes
> at the same time towards each other.
> I have tried every possible set of parameters in pull code...and i am not
> able to pull them together.
>
> it's like ...
> the DNA molecule is aligned along Y-axis and if want to pull both ends at
> the same time that means pulling in -Y and +Y direction at the same time.
>
> if anyone successfully implemented such ...simulations... Please provide me
> with some idea about this problem.
>
> Thank you
> Raghav