GROMACS version:2023
GROMACS modification: Yes/No
Hi,
I am simulating simple 8x8 lipid monolayer membranes in water and I’m seeing that the APL (calculated from Box*BoxY / 64) is decreasing over time. It seems like the box size is decreasing and the membrane is collapsing inward? Why is this the case? What are some possible fixes? My production run was done with C-rescale, semisotropic at tau=2.
What force field are you using for the lipids? I think there are some (old?) force fields that need pcoupltype = surface-tension.
I’m using a modified gromacs forcefield parametrized from ATB3.0 but after looking at it in VMD, it looks like the membrane is stretching in the z-axis, making it look more like a rectangular prism than a cube of membrane, which I suspect is decreasing the APL. Do you think this is likely an issue due to a lower tau=2? For my NPT equillibrations, I was using tau=10 so I’m perhaps thinking that the weaker coupling might be necessary.
While I think you can increase tau-p to 5 or 10, I don’t think that’s the problem here. I’m afraid that the force field parameters are the culprit. I think you should avoid generating parameters/topologies for lipids using automated tools, such as ATB. I think you should see if there are specifically generated parameters for your lipid.
