[gmx-users] bilayers move apart by nanometers upon implementing dihedral restraints on lipid tails

[Christopher Neale]

A gel-state all-trans lipid bilayer is going to be thicker than a liquid-phase lipid bilayer. 
I presume that you are starting far away from the all-trans conformation and I can imagine that this could
lead to a net force that pushes the lipids apart. Inertia might just take it from there.

You might be able to get around this by using the BD integrator, or perhaps the SD integrator with a
sufficiently strong friction. Either way, the idea behind this approach is to eliminate or reduce inertia.

Another idea, if you are simply trying to get an all-trans conformation of the lipid bilayer, you might put position 
restraints on the headgroup phosphorus atoms and slowly reduce the force constant.

Your implied idea of slowly turning up the force constant for the dihedral restraints should also work.

Chris.

khandelia hkhandelia at gmail.com 
Fri Jul 6 17:05:45 CEST 2012
Previous message: [gmx-users] Re: bilayers move apart by nanometers upon implementing dihedral restraints on lipid tails
Next message: [gmx-users] question about energy minimization
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
I am trying to change all lipid acyl tails to a trans orientation, and
thought that restraining all tail dihedrals to 180 should work quickest. I
have also tried smaller angles (120, 100) and the effect is the same,
leaflets drifting apart. 

In fact, the 180 restraint does work fine in vacuo for a single lipid. But
it is possible that the restraints are too loud. I will play with this for a
little bit. 

The dihedrals of ALL lipid acyl tails in a typical lipid bilayer are being
restrained.