[gmx-users] Membrane-protein Simulation
jalemkul at vt.edu
Mon May 6 14:26:20 CEST 2019
Please don't reply to the whole digest.
On 5/6/19 6:47 AM, Sankaran SV . wrote:
> @email on Membrane protein Simulation:
> Dear Najamuddin Memon
> Thanks for your reply. I am a little confused here. Why is the protein
> alone drifting? Typically, when we simulate protein only or protein-ligand
> systems, the protein along with the co-molecule drifts – and hence we do
> pbc correction. But in our case, we observed that the lipid does NOT move.
> Only the protein, which is supposedly embedded in the lipid, drifts.
This is not true. Everything moves. Visualize any individual lipid and
you will see it diffuse over time. Lipid diffusion timescales are
extremely long, though, so net movement is much lower. But the protein
is certainly capable of drifting (again, along with everything else) and
is simply the most prominent species that is the most obvious to have
changed its position.
> Further, when we proceeded with centering the system with respect to the
> protein (which removed the drift), and calculated the number of water
> molecules that remained in the interface layer (defined as 4.5 angstrom
> from the protein) throughout the simulation, the two numbers were
> different. That is, we obtained 16 waters molecules that remained in the
> interface layer BEFORE centering and only 5 were found to be in the
> interface layer AFTER centering. Would centering affect the number of water
> molecules that resides in the hydration layer throughout the simulation?
Centering doesn't affect anything, just visualization convention. You
need to be sure you're performing a robust calculation, but you haven't
shown us how you're doing it so it's not possible to give more precise
Justin A. Lemkul, Ph.D.
Office: 301 Fralin Hall
Lab: 303 Engel Hall
Virginia Tech Department of Biochemistry
340 West Campus Dr.
Blacksburg, VA 24061
jalemkul at vt.edu | (540) 231-3129
More information about the gromacs.org_gmx-users