Membrane-Nanoparticle simulations

GROMACS version: 2020.4
GROMACS modification: No

I am interested in studying interactions between model lipid bilayers and nanoparticles. Unbiased, all-atom MD is extremely slow, reaching around 5-10 ns/day. On the other hand, MARTINI CG is much faster to vreaching over 2 microseconds/day. Since MARTINI CG models of metallic nanoparticles may not be suitable, is there a faster way of doing AA simulations such as SMD, Umbrella sampling, etc. Which approach will be the best in such cases?


This depends very much on the questions you try to address - a good strategy is trying to identify the type of the largest energy barriers in your system and try to overcome them.

If you know what type of movement you are interested in (binding to bilayer, moving through the bilayer, etc) and can describe it with some type of reaction coordinate, enhanced sampling methods will come in handy. In that case, it might be worth having a look at the ‘AWH’ method that comes natively with GROMACS, however you will have to be able to define some type of distance, angle, etc. to look out for.

For some questions, sampling more degrees of freedom through heating up the system might help; any flavor of replica exchange simulation or hamiltonian replica exchange; though for your system, you should watch out not to melt the lipid bilayer.

If you want to know more about the nanoparticle but don’t know where to start, flooding might be a good approach to speed up nanoparticle dynamics.

Eventually, it’s a good idea to check what experimental data you have available for the system that you study and try to figure what you would need to sample to be able to compare to that data.

Hope that helps a bit,