Supposing that one molecule is static and one is dynamic,

  • can the dynamic one be solved with LINCS for its shape (angle, bond length) constraints and also keep collisions with static molecule off, using single-pass (with order-4) LINCS?
  • if single pass LINCS is not enough to keep both molecules non-interfered state and dynamic molecule's original shape, then how many passes of LINCS(order-4) would be needed on average for N atoms of dynamic molecule and M atoms of static molecule?
  • 1
    $\begingroup$ Can you (perhaps temporarily) replace LINCS by SHAKE? That way you wouldn’t have to worry about specific LINCS parameters. Also, from your question it seems that what you want is to disable non bonded interactions between the two molecules. $\endgroup$ – Juan M. Bello-Rivas Dec 15 '18 at 15:53
  • $\begingroup$ I didn't even finish understanding LINCS yet. I just need this: if an atom of a molecule is colliding with a static molecule, bump it back by solving LINCS so that no more collisions exist and also keep "rigid body" shape of dynamic molecule so that its still same molecule with some translation and rotation now. But I guess LINCS only do 1 molecule shape keeping so that it may enter half of its body into other molecule right? I'm trying to evade this. $\endgroup$ – huseyin tugrul buyukisik Dec 15 '18 at 19:54

Constrained simulation methods such as LINCS are there to keep certain degrees of freedom fixed throughout the simulation, so you can use them to make each of your molecules behave like a rigid body.

With respect to the interactions between the two molecules, I suggest you replace the non-bonded forces between the two molecules by something like a Weeks-Chandler-Andersen potential. You can do that in GROMACS using tabulated potentials.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.