James, initial heating in NVT makes sense, then equilibrate volume in NPT. I would personally use positional restraints on solute and quickly ramp temp of water and ions to 300K, then gradually relax the posutional restraints. Then let the low density system run until temperature and energies flatten at target and switch to NPT and run till volume is flat and you can rationalize the curve of rms diff over time of the fitted dilute coordinates. Then switch to NVT and run some more.
I base this on old experience and reading this list, and I'd say equilibration strategies are varied. Graph everything you can think of, and render the whole trajectory as a movie and watch it from all angles. I'm a sole fan of quickly generating a few long vacuum trajectories at low temperature at the beginning, just for the sake of fiddling with some controls on the movie, with no big investment for a few minutes or so :-).
Bill
-------- Original message --------
From: James Starlight <jmsstarlight.gmail.com>
Date:06/20/2014 11:31 AM (GMT-08:00)
To: AMBER Mailing List <amber.ambermd.org>
Subject: Re: [AMBER] On the usage of the position restrains during
equilibration
Bill,
many thanks for suggestion. But during amber workshop it has been mentioned
that system should been heated in NVT initially. In this regards should
the total workflow be
1) NVT (heating. system with applied posres on protein),
2) NPT (equilibration, system with applied posres on protein). <- what
should be provided in input file for better stabilization of box volume?
How much time water-soluble system should be equilibration? What outputs in
log should be checked (I've always find big fluctuations in reassure with
any barostat used) ?
3) NVT (production, free protein)- barostat turned off
Will such approach (production in nvt instead of npt) be suitable for
membrane protein simulation as well?
James
2014-06-20 19:36 GMT+04:00 Bill Ross <ross.cgl.ucsf.edu>:
> For what it's worth, I thought NVT is preferred for production and NPT to
> equilibrate volume. Since NVT is faster and less artificial than scaling
> the box to affect pressure.
>
> Bill
>
> -------- Original message --------
> From: James Starlight <jmsstarlight.gmail.com>
> Date:06/20/2014 7:30 AM (GMT-08:00)
> To: AMBER Mailing List <amber.ambermd.org>
> Subject: [AMBER] On the usage of the position restrains during
> equilibration
>
> Dear Amber users!
>
> >From the Amber tutorials I've learn that initial heating should be always
> done in NVT and production run in any desired ensemble. Supposing I'm
> using NPT for production run performing 100ps NVT pre-equilibration of my
> water-soluble system with applied position restrains on all atoms of my
> protein. Should I do a short NPT equilibration as well (taking all outputs
> from the previous NVT) with the such position restrains applied to the
> protein (as I've found in the Gromacs tutorials). What fluctuation to the
> system might be introduced by its initial coupling with the Monte Carlo
> barostat (for instance seriously deforming protein conformations during
> beginning of the NPT phase if pos.res on protein atoms have not been
> applied) ?
>
>
> TFH,
>
>
> James
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jun 20 2014 - 19:00:02 PDT
