[gmx-users] Unusual g_order with CHARMM36

[Ricardo O. S. Soares]

Dear Gmx users,

anyone which used the CHARMM36 (available for downloading in the Gromacs 
website) for simulating a POPC, POPE or whatever POxx membrane in 
Gromacs has found an unusual behavior in the Oleyl chain when it comes 
to the deuterium order parameter (Scd)? Let me explain a little further. I'm running 

a system with parameters consistent to those in 

[Klauda et al., J. Phys Chem. B, 2010, 114, 7830) and discussed in this thread:

http://www.mail-archive.com/gmx-users@gromacs.org/msg34641.html

Now, membrane thickness, area per lipid and the order parameter for 
the the palmitoyl chains are consistent with a liquid-crystal phase. 
Most carbons for the oleyl chains are in agreement with the literature, with 
notable exception for the central part of the
plot [C27,C28, C29], when the values form a "w" shape, instead of the 
characteristic "v" shape. I ran more than one system, each with distinct
lipid composition, they all give me these similar results.
If anyone has some ideas, I'd be grateful.

Thank you,
Ricardo.