Methyl-benzoate bonding

[vaidyanathanms]

In the parameter file for methyl-benzoate,

; id type resnr residu atom cgnr charge
1 N4a 1 MBZOA O1 1 0
2 TC5 1 MBZOA R2 2 0
3 TC5 1 MBZOA R3 3 0
4 TC5 1 MBZOA R4 4 0

[bonds]
; i j funct length force.c.
1 2 1 0.389 5000 ; cog
1 3 1 0.389 5000 ; cog

Why is there a bond between N4a and two different TC5 atoms?

Wouldn't the structure then look like

       N4a
       /.      \  
  TC5  -  TC5
       \.         /
          TC5

Instead of

     N4a
       |  
  TC5  -  TC5
       \.         /
          TC5

[ricalessandri]

Hi and thanks for the question. Yep, the structure looks like your top one. I've uploaded some more sketches to clarify further: https://github.com/ricalessandri/Martini3-mappings/blob/main/methyl-benzoate.pdf

This is because the -COOMe group is fairly rigid and this construction is very numerically stable (see, e.g., minute 11:16 of this talk).

Feel free to follow up, I'd be happy to clarify the choice further.

[vaidyanathanms]

Hi,

Thank you very much for the quick response. It makes sense now :).

Thank you,

Vaidyanathan M. S

[Clare-xiong]

I'm a little confused about the three TC5 beads. The carbon atom immediately connected to the -COOMe group (call it "alpha carbon"). Why do the TC5 map 2 non-hydrogen atoms like this :(1C_beta + 1/2C_alpha + 1/2 * C_gamma)? Why is it not like a benzene(TC5 bead map the -CH=CH- group)? When I map the molecule to the CG model, how should I deal with the benzene? Should I make the TC5 bead map like methyl-benzoate, or the benzene?

[ricalessandri]

The mapping is chosen to preserve the symmetry of the molecule. You can see for yourself by trying to draw mappings such as 1*C_alpha + 1*C_beta etc. and compare them to the recommended mapping. Note that in either case the TC5 beads would be representing 2 x C atoms, i.e., the amount of non-hydrogen atoms that they should represent. They are just placed in a more symmetric way. So, my recommendation would be to follow the 1*C_beta + 1/2C_alpha + 1/2 * C_gamma mapping so as to get a more symmetric CG model. Let me know if this clears your doubts (or not)!

[Clare-xiong]

If I understand correctly, you mean that there can be more than one way to map while guaranteeing that TC5 beads will represent two carbon atoms. The mapping method will be chosen based on considerations of symmetry. Am I right in my understanding?

[ricalessandri]

Yes, exactly.