[gmx-users] Requesting advice on pressure scaling in a protein-in-membrane system.

[Justin A. Lemkul]

Daniel Adriano Silva M wrote:
> Dear gmx users and developers,
> 
> I am facing a problem that is new for me and I will appreciate advice 
> from your expertise. I am working with a big protein complex (around 
> 100K atoms, several subunits) and I used g_membed to embed it in a 
> (aprox 24 14 nm) pre-equilibrated POPC membrane, however after embedding 
> and a 1ns of NVT equilibration I tried to proced to the NPT 
> equilibration, but I found my system exploding (things moving to fast), 
> so I tried several methods that came to my mind (like decrease the 
> P-coupling constant, decrease the timestep and some others). After few 
> tries I went back to the NVT  equilibration and took a look to the 
> pressure, for my surprise I found it to be around 1e30! So I ended with 
> a very compressed system on X-Y, I guess given the embedding of the 
> protein in the membrane. Here are the questions:
> 
> 1) Are there any reasonable simple method to reduce the pressure on the 
> protein/lipid system?
> 2) May I expect a very long time to reach the adequate pressure and 
> equilibrium given a possible change on the phase of the lipids by the 
> high pressure?
> 3) Any other useful direction?
> 
> Extra information: On this step I had frozen the Protein and excluded 
> the Protein-Protein energy.
> 

During NVT?  Turn off this freezing before proceeding, doing additional NVT if 
necessary.  Frozen groups can do weird things, and I would suspect that your 
system simply isn't stable in the presence of these frozen groups.

I have a procedure (and troubleshooting tips) for equilibrating membrane protein 
systems in my tutorial:

http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/gmx-tutorials/membrane_protein/index.html

You may find it useful.  Membrane systems do take a while to converge, but 
that's not your issue here.  You have some fundamental instability to deal with 
first.

-Justin

> Thanks,
> Daniel Silva
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================