[gmx-users] Constraining periodic cell

[David van der Spoel]

On 2010-03-01 09.49, David Chalmers wrote:
> David,
>
> I get a different simulation result if I use isotropic or anisotropic cells.
> I am not sure yet which is (more) correct, but I would like to be able to
> run both types to see.  I understand that a constrained anisotropic cell is
> not completely anisotropic, but a fully anistropic one clearly doesn't work.
>
> David

You did not say which algorithm you use either, but making the time 
constant longer might help. However, depending on your system, you may 
need substantial surface tension to keep the system neutral if you do 
use anisotropic coupling. We do not completely understand these effects 
yet, see e.g http://bugzilla.gromacs.org/show_bug.cgi?id=165 for 
additional discussions.
>
>
>>> Dear All,
>>>
>>> We are running some simulations using an anisotropic periodic cell.  We are
>>> getting 'cell runaway' with the cell becoming very long and thin.  This
>>> appears to being driven by the electrostatics of the system.  The simulation
>>> then dies because the smallest cell dimension is less than 2*cutoff.
>>>
>>> Is there a way that we can apply some constraints to the periodic cell?
>> Why not use isotropic pressure coupling?
>>
>>>
>>> Regards
>>>
>>> David
>
> ________________________________________________________________________
>
> David Chalmers                                      Lab: 9903 9110
> Faculty of Pharmacy, Monash University
> 381 Royal Pde, Parkville, Vic 3053. Australia
> David.Chalmers at pharm.monash.edu.au
> ________________________________________________________________________
>
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se