i think for refinement the dynamics are less important, since you already
are imposing restraints that modify the dynamics anyway. i often use a
collision freq of 1 for explicit solvent, and it seems to work well.
keep an eye on issues related to langevin thermostats, such as the need to
use different ig seed values in different runs, and the potential for bad
parallel scaling on high processor counds (>100) in the current code.
for constant V, taup is just ignored so it didn't hurt if you specified it,
just doesn't get used.
carlos
On Sun, Jun 28, 2009 at 4:46 AM, Sally Pias <sallypias.gmail.com> wrote:
> Hello all,
>
> I am following a protocol for explicit solvent refinement of NMR
> structures described by Xia, et al. (J Biomol NMR, 22: 317–311, 2002).
> The protocol uses Berendson temperature control throughout, with a
> heat bath coupling constant of 1.0 ps. My understanding from reading
> Amber Reflector Archive discussions is that Langevin temperature
> control is now preferred in general, particularly when one is
> interested in dynamics. Does this preference apply to NMR refinement,
> where it is not really dynamics that one is after? Is it appropriate
> to modify the protocol to use a Langevin thermostat? If so, what
> collision frequency would be suitable?
>
> I have outlined the protocol below.
>
> Thank you for any guidance,
>
> Sally Pias
>
>
> =======
> Refinement protocol, following Xia et al. (2002):
>
> 1) Gradual heating:
> 3 cycles of 40 ps (0-50 K, 50-100 K, 100-300 K)
> Constant volume
> 5 kcal/mol*A^2 restraints on protein
> TAUTP=1.0
> TAUP=0.2 (perhaps a mistake, as this is a constant volume simulation)
>
> 2) Constant pressure MD at 300 K with weak harmonic restraints
> 2 cycles of 40 ps (TAUP=0.2, then TAUP=1.0)
> 5 kcal/mol*A^2 restraints on protein
>
> 3) Constant volume MD at 300 K with weak harmonic restraints
> 1 cycle of 40 ps with 5 kcal/mol*A^2 restraints on protein
> 1 cycle of 60 ps with 0.5 kcal/mol*A^2 restraints on protein
> TAUTP=1.0
> TAUP=1.0 (perhaps a mistake, as this is a constant volume simulation)
> Time step 2 fs
>
> 4) Constant volume MD at 300 K without harmonic restraints
> 100 ps, with time step 1 fs
> TAUTP=1.0
> =======
>
> _______________________________________________
> 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 Mon Jul 06 2009 - 12:07:35 PDT
