[gmx-users] where can be obtain circled lipids bilayer?

[Mirco Wahab]

On 02.05.2013 18:32, Albert wrote:
> I've got a question about where can be obtain circled lipids bilayer?
> like shown here:
> http://wwwuser.gwdg.de/~ggroenh/membed/vesicle.png

As has already been said by others, this is not really a "circled
lipid bilayer" but rather a lipid vesicle of very very small size
(maybe 10nm-15nm diameter, guessing from the layer thickness).

There are indeed "circular lipid bilayer structures" possible, but they 
are not found very frequently. See here for example:
   Seifert, U. "Vesicles of toroidal topology. Phys Rev Lett 1991, 66 
:2404-2407
   Fourcade, B., Mutz, M., Bensimon, D. "Experimental and theoretical 
study of toroidal vesicles". Phys Rev Lett 1992, 68:2551-2554

To build a (spehrical) vesicle, you would usually start from one lipid
molecule of your choice, which can be downloaded from one of the "lipid
libraries" (e.g: http://www.nyu.edu/pages/mathmol/library/lipids/ or
elsewhere). You'd rotate that in some way, maybe oriented parallel to
the z axis of your coordinate system. After this, you write a small
script which creates points distributed on an "outer sphere" (outer
vesicle layer) and on an "inner sphere" (inner vesicle layer) separated
by the length of about two molecules. On these points, you'd copy
your single molecule, rotated by the angles of your z axis to the
direction of the point. For the inner sphere, you'd add another 180°
(wo that the lipid tails point inwards). Thats it. This structure has
to be minimized and then simulated with caution (means: a very small
time step for the first round).

M.