[gmx-users] Membrane Equilibration

[Justin Lemkul]

On 6/11/13 1:13 PM, HANNIBAL LECTER wrote:
> ref_p                    = 0.0  0.0 ??
>
> Are you sure about this??
>

That's a likely culprit.  See below for another.

>
> On Tue, Jun 11, 2013 at 12:08 PM, Dr. Vitaly Chaban <vvchaban at gmail.com>wrote:
>
>> what happens to your energy before the observed crash? does the crash
>> happen at the same time-step each time?
>>
>>
>> Dr. Vitaly Chaban
>>
>>
>> On Tue, Jun 11, 2013 at 5:22 PM, Neha <nshafique at wesleyan.edu> wrote:
>>
>>> Hi everybody,
>>>
>>> I am trying to simulate a lipid bilayer and wanted to use Parrinello
>> Rahman
>>> coupling. However, I read that Parinello-Rahman is not great for
>>> equilibration so I thought I would do a small run using the Berendsen
>>> barostat to decrease the chances of large oscillations. I then used the
>>> .gro
>>> file obtained from that run to start a new run using Parrinello-Rahman. I
>>> am
>>> not sure if that's the correct way to go about doing what I want to do.
>>>
>>> My Parrinello run also terminated a little while in, with the error that
>> 2
>>> of my atoms have moved further apart than the cut-off distance. A part of
>>> my
>>> mdp file is pasted below and I was wondering if anyone had any
>> suggestions
>>> as to stop that from happening. My time step is 0.02 ps and I thought
>>> increasing tau-p might work. I am using the compressibility from the map
>>> file on the Martini website and am worried that might be too large. Any
>>> help
>>> will be super useful! Let me know if you need any more information.
>>>
>>>
>>> ; OPTIONS FOR WEAK COUPLING ALGORITHMS =
>>> ; Temperature coupling   =
>>> tcoupl                   = nose-hoover
>>> ; Groups to couple separately =
>>> tc-grps                  = DPPC W
>>> ; Time constant (ps) and reference temperature (K) =
>>> tau_t                    = 1.0 1.0
>>> ref_t                    = 320 320
>>> nsttcouple               = 1
>>> ; Pressure coupling      =
>>> Pcoupl                   = Parrinello-Rahman
>>> Pcoupltype               = semiisotropic
>>> ; Time constant (ps), compressibility (1/bar) and reference P (bar) =
>>> tau_p                    = 5.0  5.0
>>> compressibility          = 3e-4 3e-4
>>> ref_p                    = 0.0  0.0
>>> nstpcouple               = 5
>>>
>>> ; GENERATE VELOCITIES FOR STARTUP RUN =
>>> gen_vel                  = yes
>>> gen_temp                 = 105

Here is another issue.  You're re-generating velocities (destroying the previous 
equilibration) with a barostat that doesn't usually play nice under such 
circumstances, and you're generating velocities for a temperature well below the 
target defined in ref_t.  That can cause the thermostat to go haywire, which 
then has negative effects on the pressure (which comes from kinetic energy and 
the virial), which can lead to collapse.

-Justin

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

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

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