Re: Simulated annealing error

From: Brian Radak (bradak_at_anl.gov)
Date: Thu Apr 07 2016 - 15:34:51 CDT

This is a well documented problem that was a number of known causes -
you should Google this or search in the list archives.

Quick guess - your box size is ill chosen and/or you have bad steric
clashes in your system "minimize 100" almost always fixes that for me.

On 04/07/2016 01:13 PM, faride badalkhani wrote:
> Dear all,
> I want to heat my system from 300 K to 700 K via 50 K heating steps,
> equilibrate system for a few ps and then same cooling procedure from
> 700K down to 300 K with 50 K steps. My configuration file is as follows:
>
> #############################################################
> ## JOB DESCRIPTION ##
> #############################################################
>
> # Equilibration G5 OH terminate Alkyne dendron in a Water Box
> # Parameters were taken from CHARMM 36
> # Simulated annealing
>
> #############################################################
> ## ADJUSTABLE PARAMETERS ##
> #############################################################
>
> structure Alkyne-SA/A_OH.psf
> coordinates Alkyne-SA/A_OH.pdb
>
> set temperature 298
> set outputname A_OH_SA
>
> firsttimestep 0 ;# last step of previous run
>
>
> #############################################################
> ## SIMULATION PARAMETERS ##
> #############################################################
>
> # Input
> paraTypeCharmm on
> parameters Alkyne-SA/par_all36_prot.prm
> parameters Alkyne-SA/par_all36_Alkyne.prm
> temperature $temperature
>
>
> # Force-Field Parameters
> exclude scaled1-4 ;# These are specified by CHARMM
> 1-4scaling 1.0 ;# These are specified by CHARMM
> cutoff 12.0 ;# may use smaller, maybe 10.0, with PME
> switching on
> switchdist 10.0 ;# cutoff - 2.0
> pairlistdist 14.0 ;# Promise that atom won’t move more
> than 2A in a cycle, cutoff + 2
>
>
> # Integrator Parameters
> timestep 1.0 ;# 1.0 fs/step
> rigidBonds all
> nonbondedFreq 1 ;# nonbonded forces every step
> fullElectFrequency 2 ;# PME only every other step
> stepspercycle 10 ;# redo pairlists every ten steps
>
>
> # Constant Temperature Control
> langevin on ;# langevin dynamics
> langevinDamping 5 ;# damping coefficient (gamma) of 5/ps
> langevinTemp $temperature
> langevinHydrogen off ;# don't couple langevin bath to hydrogens
>
>
> # Constant Pressure Control (variable volume)
> useGroupPressure yes ;# needed for rigidBonds
> useFlexibleCell no ;# no for water box, yes for membrane
> useConstantArea no ;# no for water box, maybe for membrane
>
>
> # Output
> outputName $outputname
>
> restartfreq 500 ;# 500 steps = every 1ps
> dcdfreq 250
> xstFreq 250
> outputEnergies 100
> outputPressure 100
> outputTiming 1000 ;# shows time per step and time to completion
>
> #############################################################
> ## EXTRA PARAMETERS ##
> #############################################################
>
>
> #############################################################
> ## EXECUTION SCRIPT ##
> #############################################################
>
> # Minimization
> #minimize 25000
> #reinitvels $temperature
>
> #simulated annealing
> for { set temperature 298 } { $temperature <= 698 } { incr temperature
> 50 } {
> run 20000
> reinitvels $temperature
> langevinTemp $temperature
> }
> run 500000
> for { set temperature 698 } { $temperature >= 298 } { incr temperature
> -50} {
> run 20000
> reinitvels $TEMP
> langevinTemp $TEMP
> }
>
>
> run 1000000 ; 1 ns
>
> But I get this error message:
>
>
> WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 288000
> PRESSURE: 288100 0 0 0 0 0 0 0 0 0
> GPRESSURE: 288100 0 0 0 0 0 0 0 0 0
> PRESSAVG: 288100 0 0 0 0 0 0 0 0 0
> GPRESSAVG: 288100 0 0 0 0 0 0 0 0 0
> ENERGY: 288100 243.0280 486.5293 172.3114
> 14.5636 334.2073 85.4743 0.0000
> 0.0000 889.1184 2225.2322 710.1980 1336.1138
> 2226.6211 687.6288
>
> ERROR: Constraint failure in RATTLE algorithm for atom 6!
> ERROR: Constraint failure; simulation has become unstable.
> ERROR: Exiting prematurely; see error messages above.
> ====================================================
>
> WallClock: 130.864319 CPUTime: 130.864304 Memory: 98.730469 MB
> [Partition 0][Node 0] End of program
>
>
> Any help will be appreciated.
>
> Regards,
> Farideh

-- 
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

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:19 CST