[gmx-users] Which membrane result is more reliable?

[Kirill Bessonov]

Hi Justin,

Sorry I forgot the attachment. Can you see if these files are ok in terms of
sequences and more or less accurately represent both DMPC and DMPE?  they
were built in Teilman lab I believe

DMPE: http://sites.google.com/site/kbessonov/dmpe.itp?attredirects=0&d=1

DMPC: http://sites.google.com/site/kbessonov/dmpc.itp?attredirects=0&d=1

(for some reason attachements do not go through mailing server)

I was trying to understand causes of helix unwinding after 100ns was due to:
---->higher packing density 50 vs 67 A^2/lipid  (I've used membrane boxes
before protein insertion on top, just to roughly estimate).
Could the +/- 15 A^2/lipid error cause this effect of instability?

--->tried to analyze edr files using g_energy and total energy option,
thinking to get potential energy of the system and conclude stability:


DMPC membrane with 62613 atoms (including water and peptide)

Energy                      Average       RMSD     Fluct.      Drift
Tot-Drift
-------------------------------------------------------------------------------
Total Energy                -891474 kJ/mol   1132.81    1131.48
-0.00114556   -190.135

T-Protein                   309.844                  25.3307    25.3304
-2.37026e-06  -0.393404


DMPE membrane with 30416 atoms (including also water and peptide)


Total Energy:
Statistics over 100000001 steps [ 0.0000 thru 200000.0156 ps ], 1 data sets
All averages are exact over 100000001 steps

Energy                      Average                   RMSD
Fluct.         Drift              Tot-Drift
-------------------------------------------------------------------------------
Total Energy                -477439 kJ/mol    839.231    790.016
-0.0049046    -980.92
T-Protein_DMPC_DMPE         309.657     3.2719    3.27189 -9.55346e-08
-0.0191069



Is it OK to multiply DMPE total energy ( -477439 kJ/mol ) by factor of 2 to
correct for atom differences between the membranes or it doesn't make sense?
I expect DMPE membrane to have slightly higher total energy due to closer
packing.....please advise.

-->I have used GridMAT, very cool program, but my results for DMPC membrane
were averaging around 80 A^2/lipid, I guess during simulation lipids drift
so the density drops (area per lipid increases) ...


What else can I do to make sure that my DMPE result is not bogus (maybe
increase are per lipid parameter and rerun  the system again?) and what
other analysis tools can I run on my two membrane systems to compare them.

Also where can I get more detailed information on g_energy, since menu
options are abbreviated and are not 100% clear?

Thank you
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20091026/c4cf9aa4/attachment.html>