[gmx-users] water with atom-based cut-off ...

[Mark Abraham]

On 4/07/2012 11:43 PM, Michael Brunsteiner wrote:
> hi,
>
> i'd like to do a simulation of a solute in water - trying to reproduce some literature data
> i try to be consistent and use an atom-based cut-off. for this each atom needs to be
> its own charge-group ...
> so when i change spc.itp
> from:
>       1  opls_116   1    SOL     OW      1      -0.82
>       2  opls_117   1    SOL    HW1      1       0.41
>       3  opls_117   1    SOL    HW2      1       0.41
> to:
>      1  opls_116   1    SOL     OW      1      -0.82
>      2  opls_117   1    SOL    HW1      2       0.41
>      3  opls_117   1    SOL    HW2      3       0.41
>
> grompp complains that for settle all atoms need to be in the same charge group.
> so i removed the settle and exclusion statements from spc.itp and instead
> put the commands

Recent versions of spc.itp files include a flexible (ie. non-settle) 
version that you could be using easily with -DFLEXIBLE.... see 
http://www.gromacs.org/Documentation/Include_File_Mechanism.

>
>
> constraints               = h-angles
> constraint_algorithm  = LINCS
> lincs_order                = 8
> lincs_iter                   = 2
>
>
> in the mdp file ... but now constraints are not applied to waters at all (O and H look as if they move independently)

h-angles converts bonds to constraints, but there have to be [bonds].

> it also didn't help renaming the water molecule, residue and atom names ...
>
> any suggestions on how i can apply atom-based cut-off with lincs or shake to water?

As you are doing, but better :-) -DFLEXIBLE and h-angles should work well.

Mark