Short answer: probably yes.
Longer: You want gamma_ln high enough to have the system temp closely track
the thermostat temperature and low enough not to strongly impact your
dynamics. From my experience, anything between 0.5 and 5 is fine for this
purpose for non-weird systems. But you dont need to rely on estimates, run
the simulation and plot target T vs. actual T and you'll see how 'tight'
your thermostatization was. If you are worried about parts of the system
not properly exchanging energy with the rest (which Id consider quite an
unlikely problem for solvated DNA), you could also plot particle velocity
distributions for different system components after the run.
Kind Regards,
Thomas
On Sat, Aug 26, 2023 at 2:25 PM Anirban Ghosh via AMBER <amber.ambermd.org>
wrote:
> Dear all,
>
> I am running a 1000ps simulated annealing with NMR restrains with my DNA
> quadruplex.
>
> My script is like below
>
> &cntrl
> nmropt=1,
> ntpr=5000,
> ntwx=5000,
> nstlim=2000000,
> dt=0.0005,
> tempi=300.0,
> temp0=0.0,
> ig=-1,
> ntt=3,
> gamma_ln=2,
> ntc=2,
> ntf=2,
> ntb=0,
> tol=0.00001,
> cut=9999.0,
> igb=2,
> saltcon=0.050
> offset=0.13
>
> I am wondering what value of collission frequency what we need to use for
> it? Does 2 is sufficient?
> Thank you for your reply.
> Best Regards,
> Anirban
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
*Dr. Thomas Steinbrecher*
Principal Scientist CADD
Roche Pharma Research and Early Development
Roche Innovation Center Basel
F. Hoffmann-La Roche Ltd
Bldg. 092/3.92
Grenzacherstrasse 124
4070 Basel
Switzerland
Phone +41 61 682 1319
mailto: thomas.steinbrecher.roche.com
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 28 2023 - 00:00:02 PDT
