Order of Minimizing/Heating/Equilibrating

From: ben estimated.com (ben_at_estimated.com)
Date: Tue Jan 07 2020 - 16:33:51 CST

Hello, I am running a MD simulation on a protein using gradual heating. My proposed method is as follows:
1. Minimize the protein at 310K
2. Reinitialize velocities and gradually heat the solvent from 0 K to 310 K while equilibrating the fixed protein, keeping the protein fixed using the Constraint feature in the NAMD config file.
3. Gradually release the constraints by running a few hundred timesteps with lower and lower ConstraintScaling values (1, 0.75, 0.5, 0.25, 0).
4. Once the ConstraintScaling value is 0 and the protein is free, reinitialize velocities and once again gradually heat the solvent from 0 K to 310 K while equilibrating for a few thousand timesteps.

Does this seem like a sound method? Most importantly: Do I need to heat the system before both fixed and non-fixed equilibrations, or only one of the two?

# Minimization
minimize 0
minimize 2500
reinitvels $temperature
run 500 ; #
constraintScaling 0.25
reinitvels $temperature
run 500 ; #
constraintScaling 0
reinitvels $temperature
run 500 ; #
minimize 500

In this example I found (https://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l.2004-2005/1310.html) that shows how to gradually lower constraints, why did they add "minimize 0" on line 2 and "minimize 500" at the last line. Is it standard to minimize the system after equilibrating it? Please let me know if there are any additions to my proposed method that I should make.

Thank you very much,

Ben Leiken

This archive was generated by hypermail 2.1.6 : Fri Dec 31 2021 - 23:17:08 CST