[gmx-users] Re: COM pulling - speptide

[Thomas Schlesier]

>
> If you're proceeding blindly, then you're likely going to frustrate yourself to
> >no end.  There is literature relevant to what you're doing - pulling on
> >peptides/proteins to unfold them is something that's been done.  So you should
> >have some methodological precedent.
>    
> >"Non-Protein" is a default group that should have been written to the .ndx file
> >by make_ndx.  The only reason I can see for this error to come up is if you've
> >removed it from the pull.ndx file.  If you're using my .mdp file, then
> >Non-Protein is a temperature coupling group.
>    
> >-Justin
>    
>  I know there is literature relevant but is there any that explains how to do it in Gromacs?

Look into Justin's Umbrella Sampling tutorial. In step 5 it explain a COM pulling simulation.

I use for this kind most of the time
pull_geometry = position
but then you must define a pulling direction (vector from 'pull_group0' to pull_group1').
For the 'pull_dim' i would set all 3 parameters to yes, if you want to do an afm-pulling experiment simulation. If you're pulling in the z-direction and have 'pull_dim = N N Y' your pull_group is free to move in the xy-plane (with no force acting on it), but if you think of an afm-experiment, there would also be a force-component in the x- and y-direction.

Greetings
Thomas