Re: Reg: deformation of water box during equilibration

From: Aron Broom (broomsday_at_gmail.com)
Date: Fri Jun 06 2014 - 22:28:02 CDT

This looks like an NVT ensemble simulation (I don't see any pressure
control). Although you might be starting with a density of 1 g/cm3, it
isn't necessarily the case that the system SHOULD have that density. In
particular, consider that forcefields don't necessarily reproduce exactly
the correct densities, but also, you have here a 35 Angstrom cubed box, but
I'm guessing you have something other than water in there? Say a protein?
 If so, your protein concentration will, out of computational necessity, be
FAR in excess of anything physically possible, such that the actual density
of the simulation should be something between water and protein. Since
protein has a solution density of ~1.35 g/cm3 this means the real density
of your system should be more than 1 g/cm3. So, once you start the
dynamics, your system should collapse to a higher density (which would lead
to some deformation, to a sphere in the extreme case, and just something
without corners in the less extreme).

For the above reasons it's often a best practice to do energy minimization,
followed by a short equilibration of the system with NPT (pressure
control), and then move on to your NVT production. It's challenging to
compute ahead of time what the density should be, so just let the dynamics
figure it out. You should run the NPT until the volume stops changing (I
think something related to this value is in the NAMD log output).

Or maybe this isn't the problem. But I see you have wrapAll on, so it's
not a visualization issue.

On Fri, Jun 6, 2014 at 10:15 PM, Akshay Bhatnagar <
akshaybhatnagar2790_at_gmail.com> wrote:

> yes i have checked. it has been mentioned that this occurs due to density
> problems. but the water box i am using is having 1.0 gm/cm3. still the
> shape is getting deformed. the configuration is below :
>
> 1-4scaling 1.0
> cutoff 12.0
> switching on
> switchdist 10.0
> pairlistdist 14.0
>
>
> # Integrator Parameters
> timestep 0.5 ;# 0.5 fs/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 off ;# don't couple langevin bath to hydrogens
>
>
> # Periodic Boundary Conditions
> cellBasisVector1 35.0 0.0 0.0
> cellBasisVector2 0.0 35.0 0.0
> cellBasisVector3 0.0 0.0 35.0
> cellOrigin 0.0 0.0 0.0
>
> wrapAll on
>
>
> # PME (for full-system periodic electrostatics)
> PME yes
> PMEGridSpacing 1.0
>
> #manual grid definition
> PMEGridSizeX 40
> PMEGridSizeY 40
> PMEGridSizeZ 40
>
> # Output
> outputName $outputname
>
> restartfreq 500 ;# 500steps = every 1ps
> dcdfreq 250
> xstFreq 250
> outputEnergies 500
> outputPressure 500
>
> run 5000000 ;# equilibration run for 10ns
>
>
>
> With Regards
> Akshay Bhatnagar
> PhD Student
> BITS Pilani Hyderabad Campus
>
>
>
> On Fri, Jun 6, 2014 at 11:07 PM, Aron Broom <broomsday_at_gmail.com> wrote:
>
>> did you search through the mailing list archives? This question is asked
>> often, and usually has the same solution.
>>
>>
>> On Fri, Jun 6, 2014 at 12:35 PM, Akshay Bhatnagar <
>> akshaybhatnagar2790_at_gmail.com> wrote:
>>
>>> Hello everyone
>>>
>>> I wanted to simulate cystine in a 30 A box but after the equilibration
>>> the
>>> water box shape is deformed. I am not understanding why it happened as i
>>> have given PME boundaries. can anyone suggest me the possible sources
>>> of errors?
>>>
>>>
>>>
>>>
>>> Please help!!
>>> With Regards
>>> Akshay Bhatnagar
>>> PhD Student
>>> BITS Pilani Hyderabad Campus
>>>
>>>
>>
>>
>> --
>> Aron Broom M.Sc
>> PhD Student
>> Department of Chemistry
>> University of Waterloo
>>
>
>

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

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:27 CST