[gmx-users] CHARMM-GUI Membrane building problems

[Björn Sommer]

> Message: 5
> Date: Fri, 6 May 2016 06:51:55 -0400
> From: Justin Lemkul<jalemkul at vt.edu>
> To:gmx-users at gromacs.org
> Subject: Re: [gmx-users] CHARMM-GUI Membrane building problems
> Message-ID:<7d29cbd8-6111-3a02-eddf-9f5e77d22756 at vt.edu>
> Content-Type: text/plain; charset=windows-1252; format=flowed
>
>
>
> On 5/5/16 11:41 AM, John Smith wrote:
>> Hello,
>>
>> I am resending this message from my Gmail account, as it appears that
>> messages sent from my Yahoo account are going to spam. Apologies if you
>> have already seen this.
>>
>> I am trying to build a simple DOPC-DPPC-Cholesterol bilayer with
>> CHARMM-GUI. The process seems to be fairly straightforward. Unfortunately,
>> there are problems with the resulting configuration of the membrane. After
>> running the equilibration scripts provided as part of the Charmm-GUI output
>> download, with the default MDP files provided, I see that several of the
>> lipids are resting in horizontal positions on the "top" of the bilayer
>> (e.g. a DOPC molecule with chains pointing sideways instead of into the
>> bilayer). So, I tried more strongly position-restraining the head-groups of
>> all three molecule types in z, and re-ran all the equilibrations from the
>> beginning. Unfortunately, I still see a couple DOPC and Cholesterol
>> molecules horizontally on the top of the bilayer. Any suggestions as to why
>> this might be, and how one might prevent it? I could weakly position
>> restrain the ends of the lipid chains in z as well, but I am not sure
>> whether this might have unwanted side effects during the equilibration.
>>
> Are the starting coordinates sane?  How quickly does the problem occur?  Any
> visual clues as to what's going on from watching the trajectories?  I have never
> seen such a thing happen, and the force required for such a displacement would
> be rather large.
>
> -Justin
>
>> Sincerely,
>> John
>>
>
Hi John, might there be a problem with the box size? Maybe CHARMM-GUI 
defined the box size in the wrong and was not able to cope with that? 
Maybe it is possible to extend the vertical size of the box a little bit.

Anyway, first you should try what Justin proposed, use e.g. VMD and have a look at the starting configuration and load the tragjectory in case the lipids are not inversely positioned right from the start. In the worst case, the packing algorithm of CHARMM-GUI does not work correctly here. Then you could use our Cm MembraneEditor (with Java GUI) or PackMol (with Scripting) to generate the membrane.

Have fun! Bjorn