[gmx-users] Insertion protein in the membrane via G_membed

[Mark Abraham]

On 20/06/2012 5:08 PM, James Starlight wrote:
> Mark,
>
> I've made changes in the input mdp file
>
> integrator     = md
> energygrps      = Protein_ADN
> freezegrps     = Protein_ADN
> freezedim      = Y Y Y
> energygrp_table
> energygrp_excl = Protein_ADN Protein_ADN
>
> here Protein_ADN is the protein_ligand defined in the index.mdp
>
> than I've processed by grompp without problems
>
> but during insertion step by follow command
> g_membed -f input.tpr -p topol.top -n index.ndx -xyinit 0.1 -xyend 1.0 
> -nxy 1000
>
> here I choose Protein_ADN as the group to be inserted and POP as the 
> group wich are membrane. Eventually I've obtained another strange error
>
> Moleculetype POP is found both in the group to insert and the rest of 
> the system.
> Because we need to exclude all interactions between the atoms in the 
> group to
> insert, the same moleculetype can not be used in both groups. Change the
> moleculetype of the molecules POP in the inserted group.
>
> I've checked my index.ndx and didt not find POP group in my first 
> Protein_ADN group. Why this error should be ?

Then it seems you're checking things that don't match each other. The 
numbers in the index file must relate to the whole system, defined by 
the [molecules] section. You need to make the index group from a file 
that corresponds to that section, i.e. from the coordinate file you give 
to grompp, which also must match that section. You should note also any 
particular requirements in the g_membed documentation about whether the 
insertion group has to be a single molecule or moleculetype.

Mark