Martini 3 bead type for azide?

[cfscosta124]

Hello,
I want to do a CG simulation of a protein containing a few non-canonical amino acids, specifically L-azidohomoalanine. I have managed to manually map out the rest of this amino acid from AA to CG following the Martini 3 bead types, but the azido group (N^-=N⁺=N-R) is trickier, since I cannot find anything where this might have already been done. Can anyone please help me?
Thank you in advance!

[ricalessandri]

As a first approximation, what I'd do is try to find logP data for methyl azide, then use that to find the regular bead that matches best that logP. You can use Supplementary Table 17 of the Martini 3 paper (Nat Methods). Then you could switch to a S-bead representing N-=N+=N. After that, as a 2nd-order correction, let's say, you may need to think whether you need an explicit dipole...

[cfscosta124]

Hi! Thank you for your quick reply.
Unfortunately, as I am finishing my PhD, I won't have the time to try this, and my advisors have since opted to go with an all-atom simulation instead, for which there are already parameters available.
However, once I'm done with the PhD in a couple months, I will try this approach you suggested.
Much appreciated!