[gmx-users] Resizing membrane (dopc, berger)

[chris.neale at utoronto.ca]

The simplest way to get around what you describe is to continue with  
your procedure and then equilibrate your membrane. Use semiisotropic  
pressure coupling and you will have what you want after 20-50 ns. You  
could alternatively build and equilibrate your own membrane, but this  
will require approximately the same amount of simulation time.

I wouldn't be surprised if you can fix this problem as a side-effect  
of the inflategro script. You could also use the make-hole version of  
gromacs 3.1.4 to keep your waters from entering the membrane during  
equilibration. However, you should be easily able to equilibrate your  
membrane within one week of wall time so the brute-force method may be  
faster for you than learning some new tools.

Chris.

-- original message --

Hello dear GROMACS users,

Lately I have been trying to user dopc membrane, using Berger united
atoms and I am confounded by a problem regarding the size of the
membrane.
The original membrane from the website is X Y Z (including 128
lipids). Now I would like to multiply the membrane by 1.5 and I cannot
find how.
Is there a (relatively) simple way of doing it?
I have tried several approaches:
1) genconf -f dopc128.gro -o dopc_new.gro -nbox 2 2 1
This option can only multiply the by *whole* integer numbers.

2) genconf -f dopc128.gro -o dopc_new.gro -box 1.5X 1.5Y Z # (X and Y
multiplied by 1.5, Z is unchanged).
The dopc_new.gro file will contain X1.5 molecules, but the water
molecules will extend much further than the lipid molecules. So now
there's a small "vacuum" in the gap between upper/lower leaflets of
the membrane in which water can invade.

I would much like to be able to use X1.5 and not X2 since that would
"save" me a whole lot of water/other molecules in the Z-axis, and
therefore simulation time too.

Thanks in advance,
-Shay