> I am facing an issue while analyzing my trajectories. After running
> equilibration under NVT (pmemd.cuda), I used cpptraj to convert the final
> .rst7 file to .pdb. The structure looks abnormal (screenshot attached).
> However, when I performed the production run under NPT (pmemd.cuda) and
> analyzed the .rst7 file with cpptraj in the same way, the output structure
> looks normal (screenshot attached).
>
> I am not sure why there is such a difference in visualization between the
> equilibration and production structures. Could this be related to imaging,
> wrapping, or my input parameters?
Input parameters or workflow. During equilibration you need NPT since the
initial coordinates likely did not have perfect density. Usually, the
initial water spacing is a bit too large between the smaller boxes that
build up the full system and also the approach to the solute is too
permissive. So as you run NVT, vacuum bubbles form since the water
collapses a bit. I would have to search the archives for my old posts on
this, but one way around is to set the closeness to something like 0.9 or
0.95 when doing solvatebox. But definately, need NPT. Also, solutes can
rotate fairly quickly (well if you are talking hundreds of nanoseconds to
microseconds) so we would generally suggest solvateoct (closer to
spherical) that the two ends of the solute do not interact across periodic
boundaries.
Dan Roe has a nice protocol --superceeding mine as usual-- that does
equilibration well. Try it out.
https://ambermd.org/tutorials/basic/tutorial18/include/Roe20.pdf
--tom
\-/ Professor Thomas E. Cheatham, III
-/- Department of Medicinal Chemistry, College of Pharmacy
/-\ Director, Research Computing and CHPC, U of Utah (on leave)
\-/
-/- tec3.utah.edu
http://www.chpc.utah.edu/~cheatham
/-\ SKH-4914 (801) 587-9652
\-/ INSCC-410 (801) 585-6318
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Sep 03 2025 - 23:30:02 PDT
