[gmx-users] building stacks with a desired system density

[Justin A. Lemkul]

Moeed wrote:
> Dear gmx users,
> 
> I tried to generate a stack of molecules using
> 
> *genconf_d -f Hexane.gro -nbox 4.0 8.0 8.0 -dist 1.1 0.55 0.55 -o 
> Hexane-stack.gro*
> 
> 
> I ran the MD simulation successfully but now I want to change the number 
> of molecules so that I get the desired system  density. To do so I 
> played a bit with -dist and -box options but I find it not easy to reach 
> desired  density. I need to have nearly 7 molecules per nm^3. 
> 
> 
> Regarding genbox program in the literature I see that  it is recommended 
> for solvation of a solute in water. (The reason I want to try this is 
> because it seems I can directly enter the number of molecules with -nmol 
> option)
> 

Where is this recommendation?  The purpose of genbox is to add solvent.  It is 
up to the user to decide what that solvent is.  For most simulations of 
proteins, that solvent is indeed water, but you are by no means required to use 
water.

You can add a specified number of molecules with -ci -nmol in tandem, not -nmol 
alone.

> 
> 
> 1-I am sorry for asking ridiculous questions but does it make sense if I 
> use genbox and I ask the program solvate a single hexane molecule with 
> hexane molecules?
> 
> genbox –cp Hexane.gro –cs HExane.gro –p Hexane.top –o solvated.pdb 
> 
> 
> (although this command line is not working I want to know if there is 
> any way to modify the command to get it working or my idea is totally 
> wrong!)
> 

You could do that, in theory.  It's useless to say "it's not working" - no one 
but you knows what that means.  Probably your box size is insufficient.

> 
> 2- Initialiy one single hexane molecule is in a box of 3*3*3 nm^3 
> (Hexane.gro). What is the best approach for doing such calculations. Can 
> I take any arbitrary box size and then based on the box size put a 
> specific No. of particles in it so that I get density?
> 

You can specify whatever initial conditions you want.  The true test of whether 
your model is correct is whether or not that density is reproduced after 
equilibration under some relevant thermodynamic ensemble.  We can impose 
whatever artificial starting configurations we want, but they are only useful if 
they last through simulation.

Your original approach (using genconf) seems fine.  Simulating that system 
appropriately may converge to the density you want, within the inherent 
limitations of the force field model chosen.  Realize that you may never be able 
to perfectly reproduce the experimentally-observed density.  Even most water 
models (that have been around for many years!) have not been able to do this.

-Justin

> 
> Thank you,
> 
> 
> 
> 
> Hexane.gro:
> 
> 
> **
> 
> Glycine aRginine prOline Methionine Alanine Cystine Serine
>    20
>     1HEX     C1    1   1.786   1.639   1.501
>     1HEX    H11    2   1.881   1.608   1.500
>     1HEX    H12    3   1.768   1.696   1.421
>     1HEX    H13    4   1.770   1.692   1.584
>     1HEX     C2    5   1.693   1.518   1.499
>     1HEX    H21    6   1.711   1.461   1.579
>     1HEX    H22    7   1.710   1.465   1.416
>     1HEX     C3    8   1.547   1.561   1.501
>     1HEX    H31    9   1.528   1.617   1.420
>     1HEX    H32   10   1.530   1.615   1.584
>     1HEX     C4   11   1.454   1.440   1.500
>     1HEX    H41   12   1.472   1.384   1.581
>     1HEX    H42   13   1.472   1.386   1.418
>     1HEX     C5   14   1.307   1.483   1.501
>     1HEX    H51   15   1.289   1.539   1.420
>     1HEX    H52   16   1.289   1.537   1.583
>     1HEX     C6   17   1.214   1.361   1.500
>     1HEX    H61   18   1.119   1.392   1.501
>     1HEX    H62   19   1.231   1.307   1.418
>     1HEX    H63   20   1.232   1.305   1.581
>    3.00000   3.00000   3.00000   0.00000   0.00000   0.00000   0.00000   
> 0.00000   0.00000
> 

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

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

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