[gmx-users] Re: problem in g_membed

[Albert]

Just a piece of advices:  you can consider equilibrate a lipids system 
which is large enough for your protein. This will save you huge amount 
of time on using tricks to add water later or enlarge the lipids.


The system MUST BE IN PBC BOX for the g_membed input coordinate, 
otherwise your job would be failed in the following step. You can 
visualize your system in VMD to double check. If it is not in the PBC 
BOX, you can use editconf -box to fix this issue.

good luck.

Albert


On 07/08/2013 11:40 AM, Justin Lemkul wrote:
>
>
> On 7/7/13 10:52 PM, pavithrakb wrote:
>> Dear Sir,
>> Thanks so much..
>> now what's the solution? Should I increase the box size?
>
> Yes.  If the protein protrudes "out" of the central image, then you 
> will have an unstable system due to atomic overlap as well as 
> violations of the minimum image convention.
>
>> Already I have centered the protein and fixed the POPE membrane size.
>> Can you tell me how to increase the box size? or is there any other
>> solution?
>
> If you need more space in the x-y plane, use genconf to replicate the 
> membrane patch appropriately.  If x-y is sufficient and only z is the 
> problem, use editconf to adjust the box size and (if desired) the 
> location of the membrane-protein complex  and then solvate with genbox 
> to fill the newly introduced void.
>
> -Justin