normal Parametrising new molecule based on known fragment - tryptophan

  • z.sichaib
  • z.sichaib's Avatar Topic Author
  • Offline
  • Senior Boarder
More
3 years 11 months ago #8504 by z.sichaib
Dear Martini users,

I am trying to parametrize a new molecule. I split the molecule in different fragments, one of them is an isoquinoline
(see: pubchem.ncbi.nlm.nih.gov/compound/Isoqui...section=2D-Structure ).

As you can see Isoquinoline is 2 planar ring molecules and I wanted to parametrize it following the way how the aminoacid tryptophane was (I needed to start from something...).

I am a little bit confused about how tryptophan is parametrized in Martini, especially which are the side chain atoms that are included inside each bead.

Based on the mapping file "trp.charmm36.map" this is what I found (I only included the heavy atoms):

[SC1]
CB CG CD1 NE1
[SC2]
NE1 CE2 CD2 CZ2 CH2
[SC3]
CG CE2 CD2 CE3 CZ3
[SC4]
CE3 CZ3 CZ2 CH2

Following this mapping, some beads can include 5 heavy atoms?

But, I also found this mapping here (page 5): www.ks.uiuc.edu/Training/Tutorials/martini/rbcg-tutorial.pdf

Can you please tell me which mapping is correct? More precisely, the side chain of TRP is represented by 4 beads. Which are the atoms included inside each bead?

One last thing, please did some of you parametrize a molecule similar to isoquinoline? Maybe you have other suggestions for me.

Thanks a lot for your help.

Zeineb

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

More
3 years 11 months ago - 3 years 11 months ago #8511 by riccardo
Regarding the mapping of TRP, see my reply in the other thread:  www.cgmartini.nl/index.php/component/kun...ptophan-mapping#8509

Regarding isoquinoline: I think it's a good idea to use a 4 x S-bead mapping for isoquinoline inspired by tryptophan in Martini 2. Maybe 3 of the 4 beads could be SC5 (used for benzene) and a more polar bead could be used for the bead describing the fragment which includes the N atom. I would validate the mapping by computing free energies of transfer and comparing them to experimental data (Pubchem reports an experimental logP for isoquinoline of 2.08: pubchem.ncbi.nlm.nih.gov/compound/Isoqui...artition-Coefficient ).
Also, make sure your bond lengths are not too short (see pubs.acs.org/doi/abs/10.1021/acs.jctc.9b00473 ) - not sure which ones are used in tryptophan but make sure they are not shorter than 0.25 nm in your isoquinoline (as a rule of thumb). More validation for the model may then depend on your specific application.

Hope this helps!
Last edit: 3 years 11 months ago by riccardo.

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

  • z.sichaib
  • z.sichaib's Avatar Topic Author
  • Offline
  • Senior Boarder
More
3 years 10 months ago #8528 by z.sichaib
Dear Riccardo,

Thanks a lot for your help and recommendation.

Yes, I am going with 4 S beads, 3 SC5 types, and 1 SNd (as a start). I am also following how TRP was parametrized and defining constraints rather than bonds. I will see if it will work of not.

Thanks a lot again.

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

Time to create page: 0.097 seconds