[gmx-users] Final state not reached in pulling simulation

[Justin A. Lemkul]

On 7/12/12 3:43 PM, lloyd riggs wrote:
> your pull force looks insanly high especially if your pulling a small piece
> of residue?  But for a whole protein of averidge 40 KDa, or 350 amino acids
> its around 2000 to 3000 from liturature (only about 6 that I could find
> anyways).  I might thus be wrong, but wounder if you have a pull rate and
> force what one wins?
>

The force constant and pull rate do not compete.  The pull rate governs how fast 
the virtual particle is pulled (at constant velocity) and the stiffness of the 
spring dictates the response of the pulled group to which it is attached.

I agree that the force constant seems excessive.  I also wonder why the pulling 
is only being applied in the z-direction.  To the OP - are the GTP and chosen 
residue completely linear in the z-direction?  Does the protein structure rotate 
over time?  If it does, then you're doing nonproductive pulling.  How are you 
measuring the distance (achieved and desired)?  Is it the z-component of the 
distance only, or the complete distance?

I would also note that one-dimensional pulling using the "distance" geometry is 
relatively inflexible.  Using "direction" or "position" geometries in concert 
with pull_vec1 and/or more sophisticated index groups may be more appropriate.

-Justin


> -------- Original-Nachricht --------
>> Datum: Thu, 12 Jul 2012 18:15:53 +0200 Von: Thomas Schlesier
>> <schlesi at uni-mainz.de> An: gmx-users at gromacs.org Betreff: [gmx-users] Final
>> state not reached in pulling simulation
>
>> It could be possible tht you do not pull into the 'right' direction. if
>> there is another group between 'GTP' and 'Residue' you will get clashes and
>> 'Residue' won't move further (could be a water molecule, or some other part
>> of 'GTP'). If this happens you should observe an increase in the force due
>> to the umbrella potential. If the problems are due to waters molecules
>> which block the 'pathway' you could just delete them. If another group is
>> in the way, you might want to change the pull-vector (and if lucky find the
>> right one). But don't know what would be the best strategy in this case.
>> Maybe you can look into what docking-people do, seems to me that your
>> simulation is related to what they do (but myself has absolute no knowledge
>> about docking simulations).
>>
>> greetings thomas
>>
>>
>>
>> Am 12.07.2012 17:26, schrieb gmx-users-request at gromacs.org:
>>> Hello all,
>>>
>>> I m performing a pulling simulation on my Protein-Mg-GTP complex. I have
>>> considered pulling between the GTP and a residue of protein. The pull
>>> code in the .mdp file im using is as follows:
>>>
>>>
>>> ; Pull code pull            = umbrella pull_geometry   = distance  ;
>>> simple distance increase pull_dim        = N N Y pull_start      = yes
>>> ; define initial COM distance>  0 pull_ngroups    = 1 pull_group0     =
>>> GTP pull_group1     = Residue pull_rate1      = -0.00005      ; 0.00005
>>> nm per ps = .05 nm per ns pull_k1         = 10000      ; kJ mol^-1
>>> nm^-2
>>>
>>>
>>>
>>> The initial distance between GTP and the residue was 7 A and the desired
>>> one was 3A. After the completion of run (10ns), I could get a trajectory
>>> where the final distance was still 4.25 A.
>>>
>>> I tried to continue the simulation for another 10ns with the same value
>>> for pull_k1 parameter and one by increasing the value to 100,000 also. In
>>> both of the case, the  trajectories showed the distance stabilized near
>>> _4.25 A only. Can anyone please tell me the reason behind it? What should
>>> I do, so that I could get the desired distance ?
>>>
>>> Any suggestion and help is welcome !!!
>>>
>>>
>>> Thanks,
>>>
>>> Neeru Sharma
>>
>> -- gmx-users mailing list    gmx-users at gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-users * Only plain text
>> messages are allowed! * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting! *
>> Please don't post (un)subscribe requests to the list. Use the www interface
>> or send it to gmx-users-request at gromacs.org. * Can't post? Read
>> http://www.gromacs.org/Support/Mailing_Lists

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

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

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