Re: Pressure in equilibration stage of MD

From: Aron Broom (broomsday_at_gmail.com)
Date: Thu Oct 18 2012 - 17:34:02 CDT

As a kind of quick answer here.

1) If you equilibrate the pressure at 100K (several hundred degrees below
the temperature of liquid water), and then fix the box size, you can only
expect that when you increase the temperature the pressure will increase.
You for sure need to do the NTP equilibration at the temperature you hope
to run the NVT simulation at, as you did the second time around.

2) You say that the water box becomes ~10 times larger when doing 300K over
100K? Is this based on the cellbasis vectors given in the .xst file, or
from visual inspection? Because if it's visual inspection, it could just
be that you haven't wrapped the water molecules and so it no longer looks
like a box.

~Aron

On Thu, Oct 18, 2012 at 5:52 PM, Mariana Graterol <marianagraterol_at_gmail.com
> wrote:

> Dear NAMD user's:
> I have a problem to balance the pressure in my system to the
> conditions of 300 K and 1 atm.
> First, I am modeling a protein using periodic boundary conditions. I
> built the box and neutralized (considering an optimal size between
> images). I have done the equilibration in stages:
> The first stage corresponds to the relaxation of the water added with
> minimization steps: 1) fix the protein (relaxing the water), 2)
> restricting the backbone of it. approximately during 30 ps.
> Second stage: It consisted in gradual heating to 100 K (in total 200
> ps, 0,0005 K heating step each step of 1 fs).
> Third stage: NPT run to adjust volume;at this point I obtained my box
> at 100 K with a suitable volume to obtain the aqueous solution density
> 1.024 g/ml.
>
> The issue arises when I want to reach 300 K, when I do it slow enough
> (I think) in a NVT run for 800 ps in total, and 0,00025 K each step,
> the system reaches the desired temperature, but rather the pressure
> increases (to approx. PRESSAVG 5000 bar).
> Even after warming up, I simulated 300 ps additional unheated, and the
> pressure does not relax in the NVT system using Langevin Control ON,
> yielding values of PRESSAVG ranging between 3000 and 6000 Bar.
>
> In trying to balance the pressure again, I made anothe NPT run, this
> time at 300 K, the pressure decreases at the expense of an increase of
> 10 times the initial volume. Which obviously does not correspond to my
> equilibrate system.
>
> I hesitated about pressure values in NAMD, are they nominal? after
> completing the protocol: Minimize-> Heat 100 K-> balance volume (NPT)
> -> Heat 300 K, it is necessary to lower these values of pressure,
> which may affect my later production run?
>
> I have read about calculating NAMD pressure, which depends on the
> virial, and therefore its fluctuations are about 1000 Bar, but I do
> not know if this increase is to be attributed to my system failure or
> disruption by increasing temperature.
>
> Thanking you for any comments, is the first time I dynamics.
> Below I include the keywords in the protocols used:
>
> Second stage (heat 100 K)
>
> # Input files
> coordinates
> C:\\Users\\serranom\\Documents\\Mariana\\eq100K\\br100p.pdb
> structure
> C:\\Users\\serranom\\Documents\\Mariana\\eq100K\\br100p.psf
> binvelocities
> C:\\Users\\serranom\\Documents\\Mariana\\eq100K\\br100p2.vel
> bincoordinates
> C:\\Users\\serranom\\Documents\\Mariana\\eq100K\\br100p2.coor
> parameters C:\\Users\\serranom\\Desktop\\parm.prm
> paraTypeCharmm on
>
> # Output files
> binaryoutput yes
> outputname C:\\Users\\serranom\\Documents\\Mariana\\eq100K\\br100
> restartfreq 1000
> restartsave yes
> binaryrestart yes
> DCDfile
> C:\\Users\\serranom\\Documents\\Mariana\\eq100K\\br100.dcd
> DCDfreq 100
> DCDUnitCell no
> outputEnergies 100
> mergeCrossterms yes
> outputMomenta 0
> outputPressure 100
> outputTiming 10000
>
> # Basic dynamics
> exclude scaled1-4
> COMmotion no
> zeroMomentum no
> dielectric 1.000000
> nonbondedScaling 1.000000
> 1-4scaling 1.000000
> vdwGeometricSigma no
> seed 12345
> rigidBonds none
>
> # PME parameters
> PME on
> PMETolerance 1.000000e-06
> PMEInterpOrder 4
> PMEGridSpacing 1.000000
> PMEGridSizeX 120
> PMEGridSizeY 120
> PMEGridSizeZ 90
> FFTWEstimate no
> FFTWUseWisdom yes
>
> # Full direct parameters
> FullDirect no
>
> # Multiple timestep parameters
> fullElectFrequency 4
> MTSAlgorithm impulse
> longSplitting c1
>
> # Harmonic constraints
> constraints on
> consexp 2
> conskcol B
> constraintScaling 1.000000
> conskfile C:\\Users\\MS\\Documents\\Mariana\\cal300\\br100p.pdb
> consref C:\\Users\\MS\\Documents\\Mariana\\cal300\\br100p.pdb
>
> # Periodic boundary conditions
> cellBasisVector1 108.447000 0.000000 0.000000
> cellBasisVector2 0.000000 111.005000 0.000000
> cellBasisVector3 0.000000 0.000000 85.008000
> cellOrigin 0.000000 0.000000 0.000000
> XSTfreq 1000
> wrapWater on
> wrapAll on
> wrapNearest off
>
> # Temperature reassignment
> reassignFreq 30
> reassignTemp 0.000
> reassignIncr 0.0015
> reassignHold 100.00
>
> Third stage (NPT run)
>
> ..
> # Timestep parameters
> numsteps 700000
> timestep 1.000000
> firsttimestep 200000
> stepspercycle 20
>
> # Basic dynamics
> exclude scaled1-4
> COMmotion no
> zeroMomentum no
> dielectric 1.000000
> nonbondedScaling 1.000000
> 1-4scaling 1.000000
> vdwGeometricSigma no
> seed 12345
> rigidBonds none
> ..
>
> # Langevin dynamics
> langevin on
> langevinTemp 100.000000
> langevinHydrogen off
> langevinDamping 5.000000
>
> # Constant Pressure Control
>
> useGroupPressure no
>
> useFlexibleCell no
>
> useConstantArea no
>
> langevinPiston on
>
> langevinPistonTarget 1.01325
>
> langevinPistonPeriod 1000
>
> langevinPistonDecay 500
>
> langevinPistonTemp 100
>
>
> Heating 300 K
>
> # Temperature reassignment
> reassignFreq 20
> reassignTemp 100.000000
> reassignIncr 0.005000
> reassignHold 300.000000
>
> NVT Run at 300 K
>
> # Langevin dynamics
> langevin on
> langevinTemp 300.000000
> langevinHydrogen off
> langevinDamping 5.000000
>
> --
> * mari *
>
>

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

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:22:10 CST