[gmx-users] Re: Need Help about GROMACS.

Justin A. Lemkul jalemkul at vt.edu
Tue May 18 16:52:00 CEST 2010



Archana S Achalere wrote:
> 
> I think the force I have applied on the two systems balance each other.
> If I calculate total force on each group:
> 
> DPPC = 734 amu, water = 18 amu
> 
>  734*128 = 93952 (This thing you calculated as 90282)

Ah yes, typo.  Sorry :)

>  18*3655 = 65790
> 
>  93952*0.07 = 6576.64
>  65790*0.1  = 6579
> 
> 
> If I reduce values of acceleration   by one order of magnitude:
> 
> ; Non-equilibrium MD stuff
>  acc-grps                 = DPPC SOL
>  accelerate               = 0.0 0.0 0.007 0.0 0.0 -0.01
> 
> and remove the center of mass motion of DPPC and water
> the simulation works well. In this case bilayer do not move along Z
> direction. But very few water can cross. So I am interested to give
> higher force so that I can observe large amount of water crossing the 
> bilayer. Which I do see but then bilayer keeps moving along bilayer normal.
> 

Well, you technically still have a force imbalance, which you can modify by the 
values of the acceleration.  As I pointed out before, you're looking to do 
something very delicate.  If you increase the magnitude of the acceleration, you 
increase the magnitude of the imbalance, thus you have a net force and a system 
that has elements colliding within it.  This is a potentially very unstable setup.

At the very least, balance the forces as best you can (-0.1 z-acceleration for 
SOL, 0.70025 for DPPC, or some other combination that works out exactly).

> 
> -archana.
> 
> On Tue, 18 May 2010, gmx-users-request at gromacs.org wrote:
> 
>>
>> Message: 1
>> From: "Justin A. Lemkul" <jalemkul at vt.edu>
>> Subject: Re: [gmx-users] Re: Need Help about GROMACS.
>> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>> Message-ID: <4BF28B9C.9040206 at vt.edu>
>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>>
>>
>>
>> Archana S Achalere wrote:
>>>
>>> Thank you Dr. Justin for the reply.
>>> Here are the details of my simulation.
>>>
>>> I am simulating  lipid bilayer (DPPC + water)  where I give equal and
>>> opposite force (along bilayer normal) on lipid and water molecules
>>> I am using 128 DPPC and 3655 water. The .pdb file is taken from
>>> Prof. Tieleman's page, University of Calgary.
>>>
>>> I am trying to calculate water flux across the bilayer.
>>>
>>> I apply the force by using nonequilibrium MD simulation option in .mdp
>>> file, where the acceleration on two groups DPPC and water is given such
>>> that the total force on each group would be equal and opposite.
>>>
>>>  I face two problems:
>>>
>>> 1. If I remove center of mass motion of whole System then the simulation
>>>  box starts suppressing  along z direction (bilayer normal) and 
>>> eventually
>>>  simulation terminates after few thousands of time steps.
>>>
>>
>> Well, you face a tenuous position.  You are forcing two groups 
>> together, and
>> unless they are delicately balanced, you'll wind up just crashing the 
>> whole
>> thing anyway.
>>
>>> 2. If I give two groups DPPC and water for center of mass motion 
>>> removal,
>>> then bilayer moves along z direction (bilayer normal).
>>>
>>
>> I would think that this is the correct method for COM motion removal, 
>> but what
>> you're seeing suggests there is a net force on the system (see below).
>>
>> <snip>
>>
>>> ; Non-equilibrium MD stuff
>>> acc-grps                 = DPPC SOL
>>> accelerate               = 0.0 0.0 0.07 0.0 0.0 -0.1
>>
>> Using these accelerations, you have a slight imbalance in your 
>> forces.  Using
>> some crude and quick calculations:
>>
>> DPPC = 734 amu, water = 18 amu
>>
>> 734*128 = 90282
>> 18*3655 = 65790
>>
>> 90282*0.07 = 6319.74
>> 65790*0.1  = 6579
>>
>> The two resulting forces should be equal in magnitude, but as you can 
>> see, they
>> are not.  The force applied by the water exceeds that applied by DPPC, 
>> so you
>> see the net motion along the z-axis.
>>
>> -Justin
>>
>>> freezegrps               =
>>> freezedim                =
>>> cos-acceleration         = 0
>>> deform                   =
>>>
>>>
>>>
>>> -Regards,
>>> Archana.
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Mon, 17 May 2010, Justin A. Lemkul wrote:
>>>
>>>>
>>>> Regardless of whether or not you've received a response, please keep
>>>> all correspondence on the gmx-users list.  I am not a private tutor.
>>>> To what post are you referring?  The list is fairly high-traffic, so
>>>> if it's been a long time since you posted, don't continue to wait for
>>>> a response.  Post a new question, expanding upon your previous problem
>>>> with any new details. You're more likely to get help if you
>>>> demonstrate you are trying to help yourself.
>>>>
>>>> You're applying opposing forces between your lipids and water?  That
>>>> certainly sounds like a recipe for collapse.  Perhaps if you can
>>>> describe (by posting to the list) what you're doing, what you're
>>>> hoping to accomplish, and what your .mdp settings are, you might get
>>>> some more useful advice.
>>>>
>>>> -Justin
>>>>
>>>> Archana S Achalere wrote:
>>>>>
>>>>> Dear  Dr. Justin Lamkul,
>>>>>
>>>>>  I apologies that I am writing directly to you the query about my
>>>>> simulation. As since long I have not got any response of my query
>>>>> posted on gmx-users.
>>>>>
>>>>> I am simulating  lipid bilayer (DPPC + water)  where I give equal and
>>>>> opposite force (along bilayer normal) on lipid and water molecules
>>>>>
>>>>> I face two problems:
>>>>>
>>>>> 1. If I remove center of mass motion of whole System then the
>>>>> simulation box stars suppressing  along z direction (bilayer normal)
>>>>> and eventually simulation terminates after few thousands of time 
>>>>> steps.
>>>>>
>>>>> 2. If I give two groups DPPC and water for center of mass motion
>>>>> removal,
>>>>> then bilayer moves along z direction (bilayer normal).
>>>>>
>>>>> Could you please give me some direction how to resolve this.
>>>>>
>>>>>
>>>>> Thank you.
>>>>>
>>>>> Regards,
>>>>>
>>>>> -Archana
>>>>> Department of Chemical Engineering,
>>>>> Indian Institute of Science,
>>>>> Bangalore, India.
>>>>>
>>>>
>>>>
>>>
>>
>>
> 

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

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

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



More information about the gromacs.org_gmx-users mailing list