[gmx-users] problem with npt

[Mark Abraham]

>
> Dear users
> I was tyring to run an npt simulation of membrane. My mdp is:

> As advised in a recent workshop on http://www.csc.fi/chem/course/gmx2007/
> i am using Parrinello-Rahman instead of berendsen. The problem is that the
> box size is getting reduced to 9.99448 9.99448 9.25077 from 10.0 10.0 10.0
> and water is compressing the bilayer.

This is the purpose of equilibration with NPT - to fix density issues. The
contraction in Z suggests that your membrane is insufficiently dense in
that direction, or has some vacuum pockets. That's not a problem in itself
so long as you equilibrate long enough. Berendsen P-coupling will likely
do the same thing.

> I also tried anisotropic coupling
> pcoupltype = anisotropic
> tc-grps = POPC SOL
> tau_t = 0.1 0.1
> ref_t = 300 300
> ref_p = 1 1 1 0 0 0
> tau_p = 10
> compressibility = 4.5e-5 4.5e-5 4.5e-5 0.0 0.0 0.0
> This time my box size is reducing to 9.99065 9.99717 9.17646.
> I cant figure out the problem. Any help will be appreciated.

This suggests the same non-problem exists.

Mark