Re: [Rdkit-discuss] SMART reaction for closing rings
Open-Source Cheminformatics and Machine Learning
Brought to you by:
glandrum
|
From: Greg L. <gre...@gm...> - 2020-06-09 15:24:54
|
Hi Shani,
If you have mapped atoms in the reactants that are not in the products,
those end up being removed
I'm not sure exactly what reaction you're trying to do, but I think you
want something like this;
rxn =
AllChem.ReactionFromSmarts("([C:1]=[C:2].[*:3][*+:4])>>[*:2]-[*:1][*+0:4][*:3]")
m1 = Chem.MolFromSmiles('C=CC([CH2+])CCC=C(C)C')
ps = rxn.RunReactants((m1,))
for p in ps: print(Chem.MolToSmiles(p[0]))
Note that I also explicitly neutralized the carbocation in the products.
Otherwise the +1 from the reactants would be carried over.
-greg
On Tue, Jun 9, 2020 at 4:42 PM Shani Levi <lev...@gm...> wrote:
> Hello,
> I'm interested in using AllChem.ReactionFromSmarts to predict product for
> a specific reaction.
> For example, I want to describe the reaction between double bonds and a
> carbo-cations.
>
> *I tried: *
> rxn = AllChem.ReactionFromSmarts("([C:1]=[C:2].[*:3][*+:4])>>[*:1][*:4]")
> m1 = Chem.MolFromSmiles('C=CC([CH2+])CCC=C(C)C')
> ps = rxn.RunReactants((m1,))
>
> *and it gave me four molecules: *
>
> [CH2+]C [CH2+]C [CH2+]CCC
> [CH2+]C(C)C
>
> the problem here that it does not describe the ring-closure molecules and
> it somehow cuts the rest of the molecule, if someone has any suggestions of
> how to change the SMARTS descriptions that it will define the right
> reaction.
>
> Thank you very much,
> Shani
> _______________________________________________
> Rdkit-discuss mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>
|