1
$\begingroup$

I have gone through this article attached, where in the methodology section they mentioned about the simulation being started with NPT ensemble keeping the box constant for 2 ns. And, I have a question regarding that, we usually start with NVT ensemble first to equilibrate and reach desired temperature for the system and then do the NPT to achieve correct density. Also, during NPT, box vector fluctuates to keep the pressure constant. Can anyone explain me.

Improve this question
$\endgroup$
4
  • 4
    $\begingroup$ The article is not very well written, but it appears as though the authors were unsure what the correct box size was. They seem to have done NPT for some steps to let the box size change, and then fix the box and do NVT for a bit to equilibrate for longer with respect to temperature, and finally switch back to NPT for the production simulation. I may have misunderstood what they did, because there are a lot of potential issues with this approach and they don't address any of them! $\endgroup$
    – Phil Hasnip
    Commented Dec 9, 2022 at 22:54
  • 2
    $\begingroup$ Thanks for the comment @PhilHasnip .... $\endgroup$
    – anupama
    Commented Dec 10, 2022 at 11:01
  • $\begingroup$ I think a book specifically about simulations will get you further and advance your knowledge more than papers. Alot of people don't know why they do the protocols they do when they publish... they just followed the herd... $\endgroup$
    – B. Kelly
    Commented Jan 14, 2023 at 23:24
  • $\begingroup$ Did the provided answer sufficiently clarify the concept that you were looking for? If yes, kindly indicate the answer as accepted. $\endgroup$
    – Hemanth Haridas
    Commented Mar 24 at 3:10

1 Answer 1

Reset to default
1
$\begingroup$

You are correct that the dimensions of the simulation box will fluctuate during the NPT run, because the system is trying to achieve correct density. If your NPT run has been performed for sufficient time (order of few nanoseconds), you must then be able to start from the final state to do an NVT run to ensure that your solvent molecules has sampled all possible configurations. The important point to keep in mind is that the convergence in NPT runs are very slow to achieve. You can refer to this paper to see how we did an NPT followed by an NVT run, and was able to reproduce experimental results.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thanks @Hemanth Haridas for the reference. I will definitely go through the article. $\endgroup$
    – anupama
    Commented Jan 10, 2023 at 4:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .