[gmx-users] Flat bottom position restraints

[Sabine Reisser]

Ok, I figured it out. With inverse FB pos res, the atoms are pushed away 
from there initial positions. Compare attached pdf, which shows the z 
coordinate of the first atom of 3 peptides, with and without the 
position restraints defined below.

Since you cannot define the direction to which the atoms/peptides are 
pushed, this doesn't help in my case.

Cheers

Sabine


Am 13.02.2015 12:14, schrieb Sabine Reisser:
> Hi,
>
> I want to use inverse FB pos res to construct a layer in the xy plane
> where my peptides are not allowed to go, to facilitate binding to the
> upper leaflet of my membrane instead of to the lower leaflet of the
> neighboring periodic image.
> In the manual it says I need three parameters
> g
> r (nm)
> k (force constant).
>
> g would be 5 because I have a layer with a normal parallel to the
> z-axis. k is also clear. r would be the thickness of the layer.
> However, when I invert the restraints I understand I would have to give
> the reference position manually (in the manual: "it is possible to
> invert the restrained region with the unrestrained region, leading to a
> potential that acts to keep the particle outside of the volume defined
> by R_i , g, and r_fb").
> How do I define the reference position R_i?
>
> Trying
> [ position_restraints ]
> 1       2       5       -1      1000
>
> doesn't give an error in grompp but I don't see how mdrun can know where
> to put the layer. Additional parameters are not allowed.
>
> I'd be thankful for a hint on this.
>
> Cheers
> Sabine
>
>


-- 
Dipl. Phys. Sabine Reißer
Karlsruhe Institute of Technology (KIT)
Institute of Physical Chemistry

Phone +49 (0) 721 / 608-45070