[gmx-users] FWD:simulationg a distance restraint with a type 6 bond in gromacs
Hans HEINDL
hheindl at terra.es
Thu Jan 7 05:48:19 CET 2010
Thanks Tom,
It may easily be that the 6- bond type works in mdrun but not in
mdrun-openmm. This would be an explanation for the ridiculous force
constants
Hans
Am Mittwoch, den 06.01.2010, 23:15 +0000 schrieb TJ Piggot:
> Please keep all correspondence on the GROMACS mailing list, it gives you a
> much greater chance of someone being able to help you.
>
> I have used this bond type before and have had no problems (with force
> constants much smaller than those which you apply). I would make sure that
> your box is big enough so that this bond is not being applied across the
> boundary. Otherwise someone else may have more ideas to help you.
>
> Tom
>
> ------------ Forwarded Message ------------
> Date: Wednesday, January 06, 2010 22:43:20 +0100
> From: Hans HEINDL <hheindl at terra.es>
> To: TJ Piggot <t.piggot at bristol.ac.uk>
> Cc:
> Subject: simulationg a distance restraint with a type 6 bond in gromacs
>
> Hi,
>
> As you remember we discussed simulation of a restraint which is not yet
> implemented in mdrun-openmm by creating a 'bond'. Type 1 bonds did not
> work but the type 6 did the job but I could not restrain the distance at
> the value I projected. The distance between the ends of my peptide
> should remain 57 angstroms and the best I accomplished was holding the
> distance fairly well for a ns and then it broke down to a much smaller
> value. (I tried to define the bond with:
>
> 1 614 6 5.761 400000
>
> with the energy term for the bond length obviously too small. But even
> if I got up to 300 000 000 up to 1 000 000 000 I got no proper result.
>
> The best results were recieved with
>
> 1 614 5 5.671 5.671 5.671 5.671
>
> but even with such an extreme example the distance tended to shrink
>
>
> kind regards
>
> Hans HEINDL
> University of Westminster UK
>
>
>
> ---------- End Forwarded Message ----------
>
>
>
> ----------------------
> TJ Piggot
> t.piggot at bristol.ac.uk
> University of Bristol, UK.
>
More information about the gromacs.org_gmx-users
mailing list