[gmx-users] Mixture of H2O and D2O

[Mark Abraham]

On 10/01/2012 10:35 PM, Hernan Ahumada wrote:
> Dear gmx users
>
> I am trying to run a MD with a mixture of D2O and H2O. I defined a new itp
> spcdeut.itp for heavy water and the usualy spc.itp for normal water. when I try
> to run grompp appear this error:
>
> "The [molecules] section of your topology specifies more than one block of
> a [moleculetype] with a [settles] block. Only one such is allowed. If you
> are trying to partition your solvent into different *groups* (e.g. for
> freezing, T-coupling, etc.) then you are using the wrong approach. Index
> files specify groups. Otherwise, you may wish to change the least-used
> block of molecules with SETTLE constraints into 3 normal constraints."

You need to follow the advice in the last sentence because of the first 
sentence.

> Any of you can help me with te meaning of this error? because if I define
> =-DFLEXIBLE I solve the error.

... because -DFLEXIBLE removes constraints, the SETTLE constraints don't 
exist, so they can't clash. You'd also be using a different water model 
than was parametrized. This can be OK for overcoming atom placement 
issues during setup, but is not useful for simulations.

Mark