[gmx-users] Annealing of shell polarizable water model

[Justin A. Lemkul]

Ivan Gladich wrote:
> Dear David
> Dear all
> I did the serial run with the same topology and grompp: even if the 
> simulation time is still short (due by the serial run), the temperature 
> profiles are the same (see attached file).
> 
> As further check, I removed the annealing and the temperature rises to 
> 160 K after ~0.5ps without problem.
> So I do not think that it is a problem of the shell polarizability with 
> the running in parallel.
> Could be a problem of the shell polarizability with annealing?
> 

You could test that relatively easily with a simple box of water.  The result 
would be useful.

Other than that, maybe the thermostat itself is causing the problem?  I always 
do annealing with a weak coupling method, not Nose-Hoover.  I've had stability 
problems with unequilibrated systems with N-H.  In principle, it shouldn't 
matter, but setting tcoupl to Berendsen and/or V-rescale would be another very 
useful diagnostic.

-Justin

> Thanks again
> Ivan
> 
> 
> On 03/22/2011 10:45 AM, Ivan Gladich wrote:
>> Yes, I am running in parallel...
>> Now I will try to run in serial to see if the problem persist....
>> Thanks
>> Ivan
>>
>> On 03/22/2011 10:00 AM, David van der Spoel wrote:
>>> On 2011-03-22 10.37, Ivan Gladich wrote:
>>>> Dear all,
>>>> I would like to heat, very slowly, a ice box of 1796 SWM4-NDP water.
>>>> This kind of water has 4 sites plus a shell and I am using a small time
>>>> step (0.1 fms) to heat my ice box from 0 K top 160K in 1 ns.
>>>> To do that I used a linear annealing from 0 to 160 K.
>>> Are you running in parallel? Unfortunately polarizable MD is broken 
>>> on more than 1 core. There is a redmine issue for this, and it will 
>>> be fixed soon.
>>>>
>>>> The simulation runs without problem but I cannot reach the desire
>>>> temperature.
>>>> In other words, if I look my md.log file I can see the ref_t that 
>>>> linear
>>>> increase from 0 to 160 K in 1 ns but the system temperature seems to do
>>>> not follow the thermostate temperature.
>>>> If I plot the temperature obtained from g_energy, the temperature of 
>>>> the
>>>> system remains constant at ~36 K.
>>>> I attach also my temperature profile up to 600ps. Due to the small time
>>>> step the simulation takes a bit of time but it is clear that the
>>>> temperature remain constant
>>>>
>>>>
>>>> I have tried to find in the mail list some similar problem without
>>>> success...
>>>> Here below I report my grompp. Maybe I missed something.
>>>> Thank in advance for any suggestions.
>>>> Ivan
>>>> #########################################3
>>>> ; VARIOUS PREPROCESSING OPTIONS
>>>> title = Ice SWM4-NDP
>>>> cpp = /usr/bin/cpp
>>>> include =
>>>> define =
>>>>
>>>> ; RUN CONTROL PARAMETERS
>>>> integrator = md
>>>> dt = 0.0001
>>>> nsteps = 14000000
>>>>
>>>> ; mode for center of mass motion removal
>>>> comm-mode = Linear
>>>> ; number of steps for center of mass motion removal
>>>> nstcomm = 1
>>>> ; group(s) for center of mass motion removal
>>>> comm-grps =
>>>>
>>>> ; OUTPUT CONTROL OPTIONS
>>>> ; Output frequency for coords (x), velocities (v) and forces (f)
>>>> nstxout = 0
>>>> nstvout = 0
>>>> nstfout = 0
>>>> ; Checkpointing helps you continue after crashes
>>>> nstcheckpoint = 10000
>>>> ; Output frequency for energies to log file and energy file
>>>> nstlog = 5000
>>>> nstenergy = 1000
>>>> ; Output frequency and precision for xtc file
>>>> nstxtcout = 10000
>>>> xtc-precision = 1000
>>>>
>>>> ; NEIGHBORSEARCHING PARAMETERS
>>>> ; nblist update frequency
>>>> nstlist = 5
>>>> ; ns algorithm (simple or grid)
>>>> ns_type = grid
>>>> ; Periodic boundary conditions: xyz (default), no (vacuum)
>>>> ; or full (infinite systems only)
>>>> pbc = xyz
>>>> ; nblist cut-off
>>>> rlist = 1.1
>>>> ;domain-decomposition =
>>>>
>>>> ; OPTIONS FOR ELECTROSTATICS AND VDW
>>>> ; Method for doing electrostatics
>>>> coulombtype = PME
>>>> rcoulomb-switch = 0
>>>> rcoulomb = 1.1
>>>> ; Method for doing Van der Waals
>>>> vdw-type = Cut-off
>>>> rvdw-switch = 0
>>>> rvdw = 1.1
>>>>
>>>> ; Apply long range dispersion corrections for Energy and Pressure
>>>> DispCorr = EnerPres
>>>>
>>>> ; Spacing for the PME/PPPM FFT grid
>>>> fourierspacing = 0.12
>>>> ; FFT grid size, when a value is 0 fourierspacing will be used
>>>> fourier_nx = 0
>>>> fourier_ny = 0
>>>> fourier_nz = 0
>>>> ; EWALD/PME/PPPM parameters
>>>> pme_order = 4
>>>> optimize_fft = no
>>>>
>>>> ;OPTIONS FOR ANNELING
>>>> annealing = single
>>>> annealing_npoints = 2
>>>> annealing_time = 0 1000
>>>> annealing_temp = 0 160
>>>>
>>>> ; OPTIONS FOR WEAK COUPLING ALGORITHMS
>>>> ; Temperature coupling
>>>> Tcoupl = Nose-hoover
>>>> ; Groups to couple separately
>>>> tc-grps = System
>>>> ; Time constant (ps) and reference temperature (K)
>>>> tau_t = 0.1
>>>> ref_t = 160.00
>>>> ; Pressure coupling
>>>> Pcoupl = Parrinello-Rahman
>>>> Pcoupltype = isotropic
>>>> ; Time constant (ps), compressibility (1/bar) and reference P (bar)
>>>> tau_p = 2.0
>>>> compressibility = 4.5E-5
>>>> ref_p = 0.0
>>>> ; Random seed for Andersen thermostat
>>>> andersen_seed = 815131
>>>>
>>>> ; GENERATE VELOCITIES FOR STARTUP RUN
>>>> gen_vel = yes
>>>> gen_temp = 0
>>>> gen_seed = 1993
>>>>
>>>> ;CONSTRAINTS
>>>> constraints = hbonds
>>>> constraint_algorithm = LINCS
>>>>
>>>
>>>
>>
>>
> 
> 
> 
> ------------------------------------------------------------------------
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================