I'm interested in quantifying the energetic contribution of non-covalent interactions between two big interfaces (a protein adsorbed on graphene oxide) from MD simulations made in GROMACS. I'm thinking about doing a lot of simulations (say 50 or more runs of 50 ns) so I need a method that gives a good compromise between speed and accuracy. The linear interaction energy method (a.k.a LIE) appears to work fine for this purpose, however , I question if this is suitable as I only find protein-ligand studies with the LIE in the literature. Thoughts and advices?
In the linear interaction energy (LIE) approach, anything can be the LIE probe: any atom or group of atoms in the simulated system.
(Yes, 50 x 50 ns is much better than 1 x 2.5 us.)
Guess you'd consider umbrella sampling (US) as a very accurate one and MMPB(GB)SA as a less accurate but easy option.
Calculation of interaction energy methods has been explained in detail in the following article:
Modeling and simulation of protein-surface interactions: achievements and challenges
https://pubmed.ncbi.nlm.nih.gov/26821792/