normal Rational for changing some sigmas

  • CharlyE
  • CharlyE's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
3 years 1 month ago - 3 years 1 month ago #8834 by CharlyE
Rational for changing some sigmas was created by CharlyE
Hello,

I would like to understand the rational behind the decrease of some sigma values from Martini 2 to Martini 3, in particular for:

- Q1-SC2: Dropped from 0.62 to 0.545
- Q1 SC3: Dropped from 0.47 to 0.44

=> Could it be that you consider Q1-SC2 as an hydrated Q1 interaction, but then it would NOT be the case for Q1-SC3 ? If this is the case, then we also have Q1-C1 interaction with Q1 considered hydrated (sigma = 0.62) and this is coherent with Q1-C3 for which the interaction is NOT considered hydrated (sigma = 0.47). My question here is not so much about the changes in sigma, than about hydration considerations, in fact.

- SC2-SC2: Dropped from 0.43 to 0.40
- SC2-SC3: Dropped from 0.43 to 0.40
- SC3-SC3: Dropped from 0.43 to 0.40

=> When I look at RDFs for hydrocarbon chains, I would tend to agree more with Martini 2 sigmas. Could you please explain the rational behind these sigmas decrease on small beads ?

Kindly,
Last edit: 3 years 1 month ago by CharlyE.

Please Log in or Create an account to join the conversation.

More
2 years 9 months ago #9043 by riccardo
Replied by riccardo on topic Rational for changing some sigmas
This reply comes very late, sorry. The rationale behind these specific choices is not easy to condense in a few lines here. Please see the Supplementary Information of the Martini 3 paper ( www.nature.com/articles/s41592-021-01098-3#Sec13 ) for a comprehensive explanation of the reparametrization and all the various choices made.

To comment briefly about the S-S sigma. It's actually slightly increased to 0.41 in the Martini 3 official release. Also, note that the S-bead size is necessarily a compromise: it needs to be able to describe 3-to-1 mapped fragments, aliphatic cyclic fragments, etc. So, the size that we converged on is the one that gave the best performance across the board. Also note that, although you can describe hydrocarbons also with S-beads, fully saturated linear hydrocarbons are best described by regular beads (sigma = 0.47 nm) (like in Martini 2).

Hope this helps!

Please Log in or Create an account to join the conversation.

Time to create page: 0.079 seconds