Re: simulation_of_big_membrane

From: Ilya Chorny (
Date: Fri Oct 26 2007 - 10:38:34 CDT

On 10/26/07, Karol Kaszuba <> wrote:
> Hello,
> I am preparing 30ns simulation of big membrane x=130, y=130.
> In properly equilibrated membrane I will insert a model of GPCR.
> However I am not sure if a membrane of this size is still enough big for
> my receptor. The problem is that my model has a long cytoplasmatic loop
> and I am not sure about the mobility of this loop - does the loop will
> escape outside the membrane.
> The simulation of the bilayer will take a lot of time so I want to be sure
> that membrane will be big enough.
> I can solve this problem in two ways:
> - increase the membrane size - I do not want to do it,
> - move the receptor much more to the corner of bilayer.
> 1) And here is my question - does the stability of a simulation will be
> influenced by the position of a protein. A picture with protein
> positioned in bilayer can be found at

The position of the protein in the plane of the membrane should have no
effect because of periodic boundary conditions. You should be able to put it
anywhere along the XY plane and get the same results. The position in the
direction normal to the membrane is important.

2) Does the time of 30ns is sufficient to equilibrate a membrane of this
> size ?

What's more important than the amount of time is some metric such as the
Area/lipid and the average P-P distance. You can also monitor the
Lipid-Lipid interaction energy. 30ns should be more than enough.

3) In the work of Kandt et al, 2007 (full title below) I found that the
> protein cavities should be solvated however in the "Building Gramicidin
> tutorial" the HOH molecules that overlaped protein were removed - if I
> understand well overlap is not equal "to water which is inside but do
> not
> overlap with protein" - in conclusion - HOH molecules should fill the
> protein cavities - correct?

Yes! You can use a program called dowser to fill cavities as well. It will
test to see if it is energetically favorable to fill the cavity.

   Christian Kandt , Walter L Ash , D Peter Tieleman
> Methods. 2007 Apr ;41 (4):475-88 17367719
> Setting up and running molecular dynamics simulations of membrane
> proteins.
> Thank you in advance,
> Regards,
> Karol

Ilya Chorny Ph.D.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:25 CST