[gmx-users] Membrane issues

[Semen Esilevsky]

> Is there some "ready to use" and easy way to build a
> nice little bilayer?

I'm working with the bilayers now and can suggest the
following procedure to build the bilayer. It is quite
simple and you will only need GROMACS and RasMol (or
any other similar visualizer). So:

1) Get a .pdb of the single lipid - either build it or
download somewhere.

2) Using the visualizer rotate it in such a way that
the tails are aligned with the Z axis. Write down the
rotation angles (let's call them rotx, roty ,rotz).

3)$ editconf -f your_lipid.pdb -o rotated_lipid.pdb
-rotate rotx roty rotz
Now you have aligned lipid.

4) Find out what area per lipid should be in your
bilayer and calculate the head-to-head distance
d_head.

5)$ genconf -f rotated_lipid.pdb -o monolayer.pdb
-nbox 8 8 1 -dist d_head d_head 0
This will produce one 8x8 monolayer

6)$ editconf -f monolayer.pdb -o monolayer2.pdb
-rotate 180 0 0
This will produce another simmetrical monolayer.

7) Determine the half of d-spacing L of your bilayer.
Now do the trick:
$ genconf -f monolayer.pdb -o temp1.pdb -nbox 1 1 2
-dist 0 0 L
$ genconf -f monolayer2.pdb -o temp2.pdb -nbox 1 1 2
-dist 0 0 L
This will produce two structures temp1 and temp2.
You'll have to use first 64 lipids from the first one
and last 64 from the second one. Copy-paste their
coordinates and just "glue" the parts together. 

8) You are done!





		
__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail