RE: gradual heating problem

From: Radak, Brian K (bradak_at_anl.gov)
Date: Mon May 09 2016 - 10:12:01 CDT

What was your initial command sequence? Something like:

temperature 0.0
minimize 100
reinitvels 5.0
run 100

should have done what you wanted.

Personally, I don't think heating is particularly necessary for many biological systems unless you have a clearly low temperature crystal structure. Even then, relaxing the system with RMSD restraints would seem to obviate the need for gradual heating (and also make possible solvent annealing at relatively high temperatures).

Brian Radak
Postdoctoral Appointee
Leadership Computing Facility
Argonne National Laboratory

9700 South Cass Avenue, Bldg. 240
Argonne, IL 60439-4854
(630) 252-8643
brian.radak_at_anl.gov
________________________________
From: dhirajks_at_gmail.com [dhirajks_at_gmail.com]
Sent: Sunday, May 08, 2016 11:21 AM
To: brian.radak_at_anl.gov
Cc: namd-l_at_ks.uiuc.edu
Subject: Re: namd-l: gradual heating problem

Thanks Brian
                       I solved the problem, i was initializing velocity by temperature using reinitvels command but i was using first time step as last time step of minimization. Setting first time step as 0 solved the problem.
However as you said you dont heat below 250 k, so do you heat your system before equilibration and after minimization. 250 and 310 k are not that different. If i am not heating below 250, do i have to heat my system at all if i want to do my simulation at 310K.

Thanks
Dhiraj

Sent from my iPhone

On May 8, 2016, at 11:04 AM, Brian Radak <bradak_at_anl.gov<mailto:bradak_at_anl.gov>> wrote:

How did you initialize velocities? The thermostat does not control initial conditions.

Also, it's very hard to hold most systems at such low temperature - normal dynamics should repartition potential energy into kinetic energy very quickly.

I stopped heating from temperatures below 250 K or so a long time ago because the system would often naturally heat up faster than my increment (it's hard to say if your choice of 100 steps will overcome this). I generally only use these kinds of protocols for annealing up to 600 K or so and then ramping down the temperature afterward.

HTH,
Brian

On 05/08/2016 10:27 AM, Dhiraj Srivastava wrote:
Hi
   I am trying to gradually heat my system after minimization and using the following set of commands.

reassignFreq 100
reassignTemp 5
reassignIncr 5
reassignHold 310

no where else in the entire configuration file i can see number 310. however after starting the calculation, system is going right away to 310 kelvin. I looked at previous post but could not find satisfactory answer. How can I resolved this issue?

Thanks
Dhiraj

--
Brian Radak
Postdoctoral Appointee
Leadership Computing Facility
Argonne National Laboratory
9700 South Cass Avenue, Bldg. 240
Argonne, IL 60439-4854
(630) 252-8643
brian.radak_at_anl.gov<mailto:brian.radak_at_anl.gov>

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:24 CST