[gmx-users] forcefields for lipids

[Christopher Neale]

Dear Shima:

I don't believe that anybody knows what lipid forcefield is best for the study of ion conduction through a membrane channel, or if it even matters. My instinct is that, for the timescales that you are likely going to use, the particular lipid forcefield that you select is not going to make a statistically significant difference, at least not any more significant than your selection of cutoffs, ion concentration, decision whether or not to apply dispersion correction, etc. That said, you certainly need a setup in which your ion channel is stable. That is where I would start. Find some studies where people have done something similar to what you want to do (if your channel is alpha helical, then focus on channels like that) and try that forcefield combination first. Before moving to umbrella sampling, check the overall stability of your system by looking at backbone RMSD and perhaps using a program like HOLE2 to ensure that the channel remains open throughout the simulation.

Chris.

-- original message --

Thanks for all your suggestions.
Honestly, I want to simulate a protein-membrane system. My chosen membrane is POPC. After the simulation of the system, I'm gonna apply the umbrella sampling on the system to study the ion conduction through the channel composed of this protein.
Before this, I had seen that the ion conduction through the gramicidin A channel in a bilayer were studied using CHARMM. Refer to that article I guess the CHARMM may give me the best  out put. Is this an acceptable choice of a force field?

 
Sincerely,
Shima