[gmx-users] Pulling two groups in opposite direction

[Justin Lemkul]

On 5/18/18 6:34 AM, Naba wrote:
>> On 5/17/18 1:39 AM, Naba wrote:
>>>> On 5/16/18 3:32 AM, Naba wrote:
>>>>> Dear all,
>>>>>
>>>>> I am using gromacs 5.1.2 and trying to pull both monomers of 128 amino
>>>>> acids from its homodimer in opposite directions along z axis. The
>>>>> interfaces of each protein chain is parallel to the z axis. I do not
>> need
>>>>> any restraints in this case. I have gone through the GROMACS manual and
>>>>> some of the previous archived messages and set the following mdp
>> options.
>>>>> ; Pull code
>>>>> pull                    = yes
>>>>> pull_ngroups            = 2
>>>>> pull_ncoords            = 2
>>>>> pull_group1_name        = chain_A
>>>>> pull_group2_name        = chain_B
>>>>> ;pull_group3_name = Protein
>>>>> pull_coord1_type = umbrella
>>>>> pull_coord2_type = umbrella
>>>>> pull_coord1_init = 0.0
>>>>> pull_coord2_init = 0.0
>>>>> pull_coord1_start = yes           ; define initial COM distance > 0
>>>>> pull_coord2_start = yes
>>>>> pull_coord1_geometry    = direction
>>>>> pull_coord2_geometry    = direction
>>>>> pull_coord1_groups      = 2 1
>>>>> pull_coord2_groups      = 1 2
>>>>> pull_coord1_dim         = N N Y
>>>>> pull_coord2_dim         = N N Y
>>>> Note that pull_coord*_dim is not relevant when using "direction"
>> geometry.
>>>>> pull_coord1_rate        = 0.01        ; 0.008 nm per ps = 8 nm per ns
>>>>> pull_coord2_rate        = 0.01         ; 0.008 nm per ps = 8 nm per ns
>>>>> pull_coord1_k           = 2000          ; kJ mol^-1 nm^-2
>>>>> pull_coord2_k           = 2000
>>>>> pull_coord1_vec        = 0.0 0.0 1.0
>>>>> pull_coord2_vec        = 0.0 0.0 -1.0
>>>>> nstcalcenergy = 1
>>>>> nhchainlength = 1
>>>>>
>>>>> But it fails to pull chain_A in positive z direction. However, chain_B
>> is
>>>>> seemed to pull in negative z direction. Someone please suggest the
>> proper
>>>>> way to pull two groups in opposite directions, or if there is anything
>>>> that
>>>>> I am missing.
>>>> What is the point of pulling in two directions? Separation of two
>>>> species requires only one reaction coordinate. For every action, there
>>>> is an equal, but opposite reaction...
>>>>
>>>> -Justin
>>>>
>>>> --
>>>> ==================================================
>>>>
>>>> Justin A. Lemkul, Ph.D.
>>>> Assistant Professor
>>>> Virginia Tech Department of Biochemistry
>>>>
>>>> 303 Engel Hall
>>>> 340 West Campus Dr.
>>>> Blacksburg, VA 24061
>>>>
>>>> jalemkul at vt.edu | (540) 231-3129
>>>> http://www.thelemkullab.com
>>>>
>>>> ==================================================
>>>
>>> Thank you Dr. Justin.
>>> Besides PMF calculations during umbrella sampling, I want to observe the
>>> interacting residues while pulling both the chain in opposite directions.
>>> In your tutorial you pulled only one chain for your specific case.
>> Whereas
>>> in my case, the point to be observed how smoothly or rigorously the
>> chains
>>> are interacting with their residues in interface due to the application
>> of
>>> two equal forces in opposite directions.
>>>
>>> Anyways, from your reply should I understand that pulling only one chain
>>> without any restraints in only one direction will do my job? Please
>> correct
>>> me the mdp settings.
>> Do not pull along only one dimension. A normal, soluble protein complex
>> will rotate in space so you should pull in all three dimensions. There
>> are several ways to accomplish this pulling, but indeed you only need
>> one reaction coordinate to induce separation.
>>
>> -Justin
>>
>> --
>> ==================================================
>>
>> Justin A. Lemkul, Ph.D.
>> Assistant Professor
>> Virginia Tech Department of Biochemistry
>>
>> 303 Engel Hall
>> 340 West Campus Dr.
>> Blacksburg, VA 24061
>>
>> jalemkul at vt.edu | (540) 231-3129
>> http://www.thelemkullab.com
>>
>> ==================================================
>>
>> Thank you again Dr. Justin.
> How do I pull only chain so that both the chains should show sliding
> movement  with respect to one another?
>
> For your convenience here is the picture of the homodimer I am dealing
> with: https://www.dropbox.com/s/6yh5vo3jcieh7rm/dummy.png?dl=0 . I want to
> pull the chains in the direction with the arrow marks given there in the
> picture.
>
> More precisely, it is N-terminal part of a very huge insoluble protein
> reported to be in dimeric form. However, comm-mode = Linear showed me
> unexpected rotation at very beginning of the run which entirely disturbed
> the initial orientation of the molecule along z axis. In contrast,
> comm-mode = Angular and pull_coord1_dim = Y Y Y with pull_ncoords = 1
> partially fulfilled my purpose but not in that way as I want both chains to
> be slide over one another.
>
> Hope you will make me more clear about all these.

For a periodic system, comm-mode = angular does not make sense.

If you want a simple sliding motion, then try pulling along only one axis.

-Justin

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

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