On Tue, Jul 14, 2020, Robert Molt wrote:
>
>I wanted to clarify a particular post from the esteemed Ross Walker from
>2011:
>
>
>"It is critical to initial heat using NVT because at low temperatures the
>calculation of pressure is inaccurate and can cause the barostat to
>overcorrect leading to instabilities. I typically either heat to 100K or so
>and then switch to NPT or I run a short (20ps or so) heating stage. The key
>is not to run for too long initially in NVT (before NPT) or you will get
>vacuum bubbles form in the solvent."
>
>My confusion is: does one not have a "discontinuity" of sorts in heating
>to 100K and then immediately beginning the NPT equilibration at T=300?
>Is the effect perhaps "small," such that it is irrelevant?
There are many ways to equilibrate. Almost all of them keep restraints on the
biomolecule + ligands, allowing only solvent to move for steps getting to a
good temperature and density. Then one can slowly relax the constraints on
the solute, before getting to a "production" run with no external restraints.
My view is that the details don't matter much, but longer is better. You
many need to do a (short) initial minimization to remove bad contacts, but
after that, there are no hard rules. Things that ran for 20 ps in 2011
would be much longer now--it can take a *long* time for solvent to become
really adjusted to the presence of the solute.
If you are concerned about some "discontinuous" temperature, run a second
simulation with different choices, and see if there are changes in the
statistical properties of the second run. (You should generally be trying
to get replicated trajectories anyway, to probe for convergence.)
Maybe others on the list can comment on problems they see with "poorer" vs
"better" equilibration protocols.
...hope this helps....dac
p.s.: a note about GPU runs: if the intial density is far from the final one,
then pmemd.cuda NPT runs can have trouble adjusting. The simplest approach is
to do the initial equilibration (just long enough to get to the target
temperature and to near the final density) on the CPU. This generally doesn't
take very long. You should still then continue "equilibration" on the GPU for
many tens of nanoseconds (depends on the size of the system and the number of
mobile ions) before removing restraints on the solute.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 14 2020 - 19:30:04 PDT
