2
$\begingroup$

I have to build a bilayer using a structure obtained in a previous MD simulation. In the structure, some chains of the molecules cross the simulation box as shown below. I added water layers to the structure as below. Since some of the structure parts are oriented out of the box, I would like to know whether doing the MD simulation for the prepared structure (below image) is correct or do I have to arrange the molecules in a manner that molecules are not going out of the box.

enter image description here

$\endgroup$

2 Answers 2

6
$\begingroup$

Usually one needs to employ periodic boundary conditions (at least in the horizontal directions). Any atoms which fly outside of the box will be mapped to the opposite side. This also has to be properly considered for the forces between the atoms, as the atoms close to your border will interact with those from the very opposite side of the simulation box.

It is best to start with one or two atoms in vaccuum to check these BC. Will the forces between them work accross boundaries? will the total momentum and kinetic energy be conserved as they pass out of the box on one side and into the box on the opposite site?

See: wiki: periodic boundary conditions

-There is also the old dos game asteroids which implements periodic boundary conditions quite nice:-)

$\endgroup$
3
  • $\begingroup$ Thank you very much for your answer. Do you think that making the box bigger would help in my case? $\endgroup$
    – Ema
    Sep 13, 2023 at 14:25
  • 1
    $\begingroup$ I don't know enough about your application. The best way forward would be to run a set of simulations, meausure the desired quantities and see if they stil vary with box size. $\endgroup$
    – MPIchael
    Sep 13, 2023 at 14:36
  • 3
    $\begingroup$ Also, if you are new to MD Simulations I'd recommend to start out with a simpler problem to verify your BC and timestepping. Can you reproduce ideal gas properties etc? $\endgroup$
    – MPIchael
    Sep 13, 2023 at 14:37
3
$\begingroup$

The biggest issue in your subsequent simulation won't be the periodicity of the box -- it will be trying to equilibrate your system down from what looks like a tangle of lipids into an orderly structure. That can happen in a simulation, but (1) it may not, depending on initial condition (2) it's wasting computer time to arrive at a foregone conclusion.

Most lipid bilayer simulations use a software like Packmol to pre-arrange the lipid molecules into a suitable bilayer.

$\endgroup$
2
  • $\begingroup$ Thank you very much for your answer. The problem I have is I need to use a previously equilibrated structure to set up my system. In the pre-equilibrated structure, some chains are going out of the box, and I cannot change the arrangement of molecules in that one. $\endgroup$
    – Ema
    Sep 17, 2023 at 10:16
  • $\begingroup$ "I need to use a previously equilibrated structure" -- but why that structure? Or in different terms, what exactly is that structure an equilibration of / in equilibrium with? I am not sure what phenomenon are you hoping to observe (in under one microsecond!) starting from that structure. Such drastic rearrangements are rare in MD, and so it is to be expected that someone trying to do this in MD will run into all kinds of problems that other MD practitioners have not tried to find answers to. $\endgroup$ Sep 18, 2023 at 11:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.