[gmx-users] Position restraints problem

[Mark Abraham]

On 28/03/2012 6:52 PM, Jernej Zidar wrote:
> Hi.
>    After successfully importing a CHARMM-generated PDB file to GROMACS
> I set out to do some short simulations. While all calculations
> finished without a problem if everything but the water molecules were
> fixed.
>
>    Removing the position restraints led to the system blowing up. Using
> J. Lemkul's tutorial suggestion I generated an index file selecting
> one lipid atom from sphingomyelin and one from cholesterol using
> make_ndx and the selection: a P | a O3 (alternatively I tried also "6
> | 7 | 17 | 18 | a P | a O3"). Any of the two selections selects only
> atoms from the lipid part of the system.
>
>    After creating the index file I used genrestr to position restrain
> the movements of the selected atoms and obtain a position restrain ITP
> file. I included this ITP file at the end of the lipid topology ITP
> file as instructed on GROMACS' website.
>
>   In the last step I edited the MDP file to use the position restraints
> file for some lipid atoms but something went wrong as GROMPP
> complains:
> Fatal error:
> [ file sys9-tmp-ions-mini-nofix-2-posre.itp, line 116 ]:
> Atom index (12014) in position_restraints out of bounds (1-11954).
> This probably means that you have inserted topology section
> "position_restraints"
> in a part belonging to a different molecule than you intended to.
> In that case move the "position_restraints" section to the right molecule.
> For more information and tips for troubleshooting, please check the GROMACS
> website at http://www.gromacs.org/Documentation/Errors
>
>    Indeed, checking the lipid ITP file revealed there is no atom index
> 12014, yet there is an atom index 12014 (with the proper name) in the
> GRO file (that I used to create both the index and the restraints
> file) where all the atoms are listed. What have I done wrong?

See the warning in genrestr -h.

If all you're doing is adding a single atom of position restraint per 
moleculetype, you can do that by hand faster than using make_ndx and 
genrestr and adding the #include.

Mark