From: Allan Haldane (allan.haldane_at_mail.mcgill.ca)
Date: Wed Nov 15 2006 - 21:18:13 CST
I finally got a chance to try out your suggestions for the pressure
profile, and they made a big improvement. It's still not quite right,
I stepped up to 120 lipids with 2 ns equilibration, and 10 ns profile,
and fixed the head groups during heating (I think this last one was the key).
I did not randomly rotate the lipids from their position in the vmd
template though, and I heated very quicky (20 ps)
Here's the lateral pressure profile (Pz - (Px+Py)/2):
I now get the two big and two little peaks, as seen in some of Sculten
et al's results. (though not run D4, the most similar to my setup) The
center of the membrane should be at a slightly positive pressure,
though, and overall the profile is too negative.
I understand the positive pressure there is due to entropy effects,
so I guess I should do random rotations to disorder it. Maybe I should
just build the membrane from scratch and forget the vmd membrane
generator? (it also seems to have too high area per lipid) Also, in the
future I will heat over 500 ps, I'm not sure how much of an effect the
short heating had here. Unless someone can see some other error in the
pressure profile, I'll try creating my own membrane.
Quoting Marcos Sotomayor <sotomayo_at_ks.uiuc.edu>:
> The membrane plugin of VMD provides membrane templates that are a bit
> too ordered. I would suggest you do what Justin did during
> "After this initial construction stage, the headgroup atoms were
> fixed and the rest of the atoms heated to 300 K for 500 ps to "melt"
> the aliphatic carbon tails. Fixing the headgroups prevented
> oppositely charged groups in neighboring lipids from bonding with
> each other, a condition that would prevent proper solvation of the
> interface. "
> Gullingsrud/Schulten (Biophysical Journal 86:3496-3509 (2004))
> Fixing the headgroups during part of the equilibration may help you
> get a more disordered (real) membrane.
> In addition, note that all the averages presented in Justin's paper
> are performed over simulation with at least 1.5ns of equilibration
> and then 10~ns of dynamics. Also, how many lipids do you have in your
> membrane? 60? Seems a bit small...
> PS: Why do you have PmeGridsizeZ 128 ? Is your system that large in
> the Z direction? You may also want to use some of the NAMD features
> to eliminate center of mass motion.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:11 CST