[gmx-users] set up self assembly for mixture of lipids using genbox

[Mark Abraham]

maria goranovic wrote:
> Hello
> 
> I am trying to using genbox to set up a random 2-lipid mixture. Is there 
> a way to do this? First, i put 100 POPC molecules randomly in the sim. 
> box. Now I want to add 28 POPS lipids. How can I do this ?

I'm not aware of a direct method for doing this. Creating interstices 
for arbitrarily shaped molecules is non-trivial.

One approach is to choose your lipid mixture and to concatenate 
cunningly-sized boxes each with a pure sample of one component of the 
mixture. Then a long equilibration will see the mixtures randomize. You 
can probably get away with a long "pseudo-equilibration" period with a 
very cheap physical model to make sure you get away from the highly 
ordered state, then a shorter "real" equilibration to move into the 
right ensemble from that intermediate point.

Mark