Re: reassignTemp question

From: Martin, Erik W (Erik.Martin_at_stjude.org)
Date: Tue Oct 29 2013 - 18:55:08 CDT

Thank you, This is definitely helpful. Actually, I was originally thinking of doing it like this but thought that "reassignTemp" was the way to do this type of thing these days. I end up getting several nanoseconds a day so it would be kind of a nuisance to simply restart at a new temperature, although I suppose I could do that to through a shell script. It seems that in the configuration file would be easier though.

Thanks again,
Erik

From: Aron Broom <broomsday_at_gmail.com<mailto:broomsday_at_gmail.com>>
Date: Tue, 29 Oct 2013 16:44:52 -0500
To: Ajasja Ljubetiè <ajasja.ljubetic_at_gmail.com<mailto:ajasja.ljubetic_at_gmail.com>>
Cc: Erik Martin <erik.martin_at_stjude.org<mailto:erik.martin_at_stjude.org>>, "namd-l_at_ks.uiuc.edu<mailto:namd-l_at_ks.uiuc.edu>" <namd-l_at_ks.uiuc.edu<mailto:namd-l_at_ks.uiuc.edu>>
Subject: Re: namd-l: reassignTemp question

ahh yes that's right. Thanks for the correction.

On Tue, Oct 29, 2013 at 5:26 PM, Ajasja Ljubetiè <ajasja.ljubetic_at_gmail.com<mailto:ajasja.ljubetic_at_gmail.com>> wrote:
Actually you create a loop wherein you call

langevinTemp $myTemp
LangevinPistonTemp $myTemp; #(if you are using a constant pressure as well)
run $mySteps
# increase myTemp and loop

temperature just reassigns the velocities of all the atoms (and is used for initialization).
But for 1ns or 2ns long segments I would personally use separate configuration files, which also makes it easier to restart a simulation. (Depends on how long it takes to calculate the said nanosecond).

Best regard,
Ajasja

On 29 October 2013 21:40, Aron Broom <broomsday_at_gmail.com<mailto:broomsday_at_gmail.com>> wrote:
you can do this with the langevin thermostat. I don't recall exactly, but I've done it before, the extra lines you need in whatever temperature loop you are iterating over look something like this:

"set currentTemperature [whateveryouwant]"
"temperature $currentTemperature"

In fact, I'm not sure that you actually want to use "reassignTemp" if you want a really gradual change, I think you might just want to use something like those lines above. I could have the specifics wrong, but it is doable with minimum work.

On Tue, Oct 29, 2013 at 4:26 PM, Martin, Erik W <Erik.Martin_at_stjude.org<mailto:Erik.Martin_at_stjude.org>> wrote:

Hi, I was wanting to gradually increase the temperature of my system, which seems easy enough with the reassignTemp options. However, I have a quick methodological question. How do you then use a thermostat to control that temperature during the increment that its unchanged? This does not reset the Langevin temperature. Is there another type of thermostat thats commonly used? To be clear, I want to do this very slowly and want the temperature to be stable over about 2ns of simulation before changing.

Thanks a lot,
Erik

________________________________
Email Disclaimer: www.stjude.org/emaildisclaimer<http://www.stjude.org/emaildisclaimer>
Consultation Disclaimer: www.stjude.org/consultationdisclaimer<http://www.stjude.org/consultationdisclaimer>

--
Aron Broom M.Sc
PhD Student
Department of Chemistry
University of Waterloo
--
Aron Broom M.Sc
PhD Student
Department of Chemistry
University of Waterloo

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:54 CST