From: Hardy, David Joseph (dhardy_at_illinois.edu)
Date: Tue Sep 13 2016 - 14:13:52 CDT
Dear Jack,
You are mixing various protocols in your simulation that don't make sense together. You have constant temperature and constant pressure running together with your heating protocol. First, you should get rid of the thermostat and barostat, as they are working against your heating protocol. Also, the fatal error you are seeing is likely due to the barostat, so that should go away. Second, your reassignFreq and reassignIncr choices are too small. Keep in mind that a temperature reassignment is setting random velocities (to that temperature) for all of your atoms, but this suggests that your system will require some number of steps to "settle down." More typically, you would change the temperature using a bigger increment and allow the system to equilibrate to that new temperature over, say, 100 to 1000 steps. So as a smallest increment, try reassignFreq 200 and reassignIncr 1, or maybe better try larger increments of reassignFreq 1000 and reassignIncr 5.
Best regards,
Dave
--
David J. Hardy, Ph.D.
Theoretical and Computational Biophysics
Beckman Institute, University of Illinois
dhardy_at_illinois.edu<mailto:dhardy_at_illinois.edu>
http://www.ks.uiuc.edu/~dhardy/
On Sep 13, 2016, at 8:34 AM, Jackson Cavett <jcavett_at_mail.bradley.edu<mailto:jcavett_at_mail.bradley.edu>> wrote:
Hello NAMD Users,
I am attempting to heat a molecule from 10K to 310K over the course of a few simulations. I put in the following parameters to accomplish this heating:
reassignFreq 2
reassignTemp 10
reassignIncr 0.01
reassignHold 60
However, when I run the simulation, I get the periodic boundary error on the first timestep. Here is the log file which shows the problem:
REINITIALIZING VELOCITIES AT STEP 100000 TO 10 KELVIN.
TCL: Running for 10000 steps
REASSIGNING VELOCITIES AT STEP 100000 TO 60 KELVIN.
PRESSURE: 100000 110.253 -0.316211 -6.80342 -0.316211 126.441 1.37312 -6.80342 1.37312 145.667
GPRESSURE: 100000 116.487 -0.516909 -5.56659 0.0868719 131.886 1.15381 -5.4699 1.44817 149.358
ETITLE...
OPENING EXTENDED SYSTEM TRAJECTORY FILE
REASSIGNING VELOCITIES AT STEP 100002 TO 60 KELVIN.
FATAL ERROR: Periodic cell has become too small for original patch grid!
Possible solutions are to restart from a recent checkpoint,
increase margin, or disable useFlexibleCell for liquid simulation.
I don't understand why the temperature is being reassigned to 60K when I stated that it should be reassigned to 10K every two timesteps. Can anyone explain what is happening here?
Here is the full configuration file I used. Please note what I did with the binVelocities. I'm not sure if I can comment them out like that when I use a restart file, so please let me know what I should do with that. The reason I comment it out is that I want to set the temperature of my simulation while still using a restart file.
#############################################################
## JOB DESCRIPTION ##
#############################################################
# Equilibration of alphVbeta3 after minimization
# Shorter simulations so dcd can be analyzed
#############################################################
## ADJUSTABLE PARAMETERS ##
#############################################################
structure ../alphaVbeta6_wb.psf
coordinates ../alphaVbeta6_wb.pdb
set temperature 10
set outputname beta6_equil_1
# Continuing a job from the restart files
if {1} {
set inputname ../minimization/beta6_min_fixed_protein
binCoordinates $inputname.restart.coor
#binVelocities $inputname.restart.vel
extendedSystem $inputname.xsc
}
firsttimestep 100000
#############################################################
## SIMULATION PARAMETERS ##
#############################################################
# Input
paraTypeCharmm on
parameters ../par_all27_prot_lipid.inp
temperature $temperature
# Force-Field Parameters
exclude scaled1-4
1-4scaling 1.0
cutoff 12.0
switching on
switchdist 10.0
pairlistdist 14.0
# Integrator Parameters
timestep 2.0 ;# 2fs/step
rigidBonds all ;# needed for 2fs steps
nonbondedFreq 1
fullElectFrequency 2
stepspercycle 10
# Constant Temperature Control
langevin on ;# do langevin dynamics
langevinDamping 1. ;# damping coefficient (gamma) of 1/ps
langevinTemp $temperature
langevinHydrogen no ;# don't couple langevin bath to hydrogens
# Periodic Boundary Conditions
cellBasisVector1 121 0.0 0.0
cellBasisVector2 0.0 93 0.0
cellBasisVector3 0.0 0.0 57.5
cellOrigin 20.045 -30.903 25.591
wrapWater on
wrapAll on
wrapNearest off
# PME (for full-system periodic electrostatics)
PME yes
PMEGridSpacing 1.0
#manual grid definition
#PMEGridSizeX 45
#PMEGridSizeY 45
#PMEGridSizeZ 48
# Constant Pressure Control (variable volume)
useGroupPressure yes ;# needed for rigidBonds
useFlexibleCell no
useConstantArea no
langevinPiston on
langevinPistonTarget 1.01325 ;# in bar -> 1 atm
langevinPistonPeriod 100.0
langevinPistonDecay 50.0
langevinPistonTemp $temperature
# Output
outputName $outputname
restartfreq 1000 ;# 500steps = every 1ps
dcdfreq 500
xstFreq 500
outputEnergies 200
outputPressure 200
#############################################################
## EXTRA PARAMETERS ##
#############################################################
#############################################################
## EXECUTION SCRIPT ##
#############################################################
#For heating, use temperature reassignment
reassignFreq 2
reassignTemp 10
reassignIncr 0.01
reassignHold 60
# Minimization
#minimize 10000
#reinitvels $temperature
reinitvels $temperature
run 10000 ;#
Thank you all for your help!
Jack
This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:43 CST