GROMACS version: 2023.3
GROMACS modification: No
Hello,
I ran 10 x 10ns production steps on my protein in a membrane system prepared using CHARMM-GUI. I used all the default settings except for increasing the time of a single production run to 10ns instead of the default 1ns. So in total 100ns.
I am not very experienced in molecular dynamics (MD) and wonder if the results obtained are reliable in terms of the protein’s stability within the membrane. Can the outcome be used for further docking studies of the ligand into such protein?
Thanks for the feedback and suggestions!
10 ns isn’t enough long enough for the lipids to relax, let alone the protein dynamics. I would suggest 100 ns or more. RMSD, Rg, and RMSF won’t tell you anything about stability, necessarily. You have to specifically investigate structural properties of your protein that are expected to impact ligand binding or conformational change.
1 Like
I might have written it confusingly… This is, of course, 100 ns — 10 runs, each 10 ns, so 100 ns in total
Sure, 100 ns is more reasonable, but you still need to do the type of analysis I mentioned above. The boilerplate stuff won’t tell you much.
1 Like
Great. My problem, though, is more complex. The protein I am searching for, in the organism it is bound to the membrane. Moreover, the crystallographic data misses a long part of the single coil.
I have compared the AlphaFold-generated structure with all the available crystal structures and confirmed they overlap very well. That’s why I decided to stick with the AlphaFold structure, anchor it in the membrane, run MD, and now I will redock known ligands (co-crystalized). If successful, I will dock the one I’m really interested in and run - once again - 100ns of MD to ensure its stability.
Hey, @jalemkul
One more question, if you don’t mind. Does 100 ns of simulation, including the docked ligand, assume its stability within the binding pocket if it doesn’t leave after that time? Or should I also consider whether there’s actually a pore it could escape through?
Thanks in advance.
I’m not sure I follow the question entirely, but I certainly wouldn’t make any argument about stability (which is really a thermodynamic concept) based on 100 ns of simulation, which is still relatively short by modern standards. And many ligands may have microsecond or millisecond off-rates.