[gmx-users] triclinic option

[Erik Lindahl]

Makoto Yoneya wrote:

>Dear GROMACS peoples:
>
>I'd like to use triclinic option in GROMACS-3.0.5.
>With grompp.mdp like,
>; Pressure coupling
>; Pcoupl              =  Parrinello-Rahman
>Pcoupl              =  berendsen
>pcoupltype          =  triclinic
>tau_p               =  2.0 2.0 2.0 2.0 2.0 2.0
>compressibility     =  4.5e-5 4.5e-5 4.5e-5 4.5e-5 4.5e-5 4.5e-5
>ref_p               =  1.0 1.0 1.0 0.0 0.0 0.0
>I'd got the following error message from grompp_d (in both berendsen and
>Parrinello-Rahman).
>
>ERROR: invalid enum 'triclinic' for variable Pcoupltype, using 'Isotropic'
>Next time use one of: 'Isotropic' 'Semiisotropic' 'Anisotropic'
>'Surface-Tension'
>
>What are prerequisites to use the triclinic option?
>
It depends on what you mean by 'triclinic' :-)

First, Gromacs always works with triclinic cells, so you don't have to
say that in the input files. If you want to specify a triclinic box in a
.gro file you should use a last line with the format:

box_xx box_yy box_zz box_xy box_xz box_yz

So, a 'normal' rectangular with three numbers just implies that the
off-diagonal elements are zero.

The other possible 'triclinic' part is the pressure coupling where we
change the box shape. Just select 'anisotropic' pressure scaling, and if
the off-diagonal elements of your compressibility are non-zero the box
will be able change shape and not only size. The classic anisotropic but
shape-preserving pressure scaling is achieved by setting the
off-diagonal compressibilities to zero, e.g. by only specifying three
values.


Cheers,

Erik