[gmx-users] Re: LINCS metal-ligand bonds

[Justin Lemkul]

On 10/12/12 10:39 AM, Dr. Vitaly Chaban wrote:
>>
>> On 10/12/12 2:42 AM, tarak karmakar wrote:
>>> Dear ALL,
>>>
>>> In my protein I need to constraint the length between the Metal and
>>> the N of Histidine residues.  In the .rtp file I didn't specify this
>>> Mn-N connectivity; these are more or less coordinate bonds.  So to do
>>> that I specified the Mn, N-atom indices  and the corresponding lengths
>>> in the topology file, part of it is given below
>>>
>>>
>>> [ constraints ]
>>> ;  index1  index2   funct          length(nm)
>>>       6062   6063       1        0.11490000000
>>>       6062   6064       1        0.11490000000
>>>       6060   4309       1        0.21000000000
>>>       6060   4268       1        0.22700000000
>>>       6060   4979       1        0.22600000000
>>>
>>>
>>>    In the .mdp file I have included the following lines
>>>
>>>
>>> constraints             = h-bonds       ; covalent h-bonds constraints
>>> constraint_algorithm    = LINCS         ; LINear Constraint Solver
>>> continuation            = yes           ; apply constraints to the
>>> start configuration
>>> lincs_order             = 4             ; highest order in the
>>> expansion of the contraint coupling matrix
>>> lincs_iter              = 1             ; number of iterations to
>>> correct for rotational lengthening
>>> lincs_warnangle         = 30            ; [degrees] maximum angle that
>>> a bond can rotate before LINCS will complain
>>>
>>>
>>> [ I followed the above mentioned protocol and I see Mn-N distances are
>>> not changing during the simulation, so served my purpose.]
>>>
>>> So for all sorts of specified constraints (H-covalent bonds, Mn-N
>>> coordinate bonds), MD will make use of LINCS algorithm !! . But my
>>> query is about whether LINCS will be equally applicable as that of for
>>> covalent bonds in constraining Mn-N distances where the real(?)
>>> bonds(connectivity) don't exist !!
>
> Introducing constraints between non-covalently bound sites may (and
> often does in practice) result in the instability of your system and
> may require smaller time-step.
>
> Personally, I would introduce morse potential with a weak force
> constant for the same task .
>

For what it's worth, I've employed a similar setup as the OP and it worked quite 
well, though there are probably several ways to address this issue.  I can think 
of at least four.

-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

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