Re: [Rdkit-discuss] bond order is lost in reaction transformation

[Devendra D. <dev...@gm...>]

Hi Paolo,
Thanks for solution, but this didn't work for me.

Let me explain the problem again. I have following reaction where hydrogen
migrate from nitrogen to carbon by 1,5 H-shift in forward direction. For
this reaction, I have following SMIRKS
"[O,NX3;!H0:2][cX3z2;r5:3]:[c;r5:4][c;r5:5]:[cr5R{1-2}:6]>>[O,S,NX2:2]=[CX3;z2;r5:3][C;r5:4]=[C;r5:5][CX3;r5;!H0:6]"
[image: image.png]
When I applied following code to input smiles "CN1C(NC2=NC=CC=C2)=CC=C1"

reactant1 = Chem.MolFromSmiles('CN1C(NC2=NC=CC=C2)=CC=C1')
rxn =
AllChem.ReactionFromSmarts('[O,NX3;!H0:2][cX3z2;r5:3]:[c;r5:4][c;r5:5]:[cr5R{1-2}:6]>>[O,S,NX2:2]=[CX3;z2;r5:3][C;r5:4]=[C;r5:5][CX3;r5;!H0:6]')
ps=rxn.RunReactants((reactant1,))
print(len(ps))

print(Chem.MolToSmiles(ps[0][0]))

I got unusual product where bond order in five membered ring is not
well-behaved. (Nitrogen and Carbon have five valencies)
[image: image.png]

Thanks

Devendra

On Sat, Jan 4, 2020 at 2:54 AM Paolo Tosco <pao...@gm...>
wrote:

> Hi Devendra,
>
> Your starting molecule is aromatic, as is your product; by default RDKit
> writes aromatic SMILES.
> If you wish to write your molecule as kekulized SMILES, do the following:
>
> In [1]: from rdkit import Chem
>
> In [2]: smi = "CN1C(NC2=NC=CC=C2)=CC=C1"
>
> In [3]: mol = Chem.MolFromSmiles(smi)
>
> In [4]: Chem.MolToSmiles(mol)
>
> Out[4]: 'Cn1cccc1Nc1ccccn1'
>
> In [6]: Chem.Kekulize(mol)
>
> In [7]: Chem.MolToSmiles(mol, kekuleSmiles=True)
>
> Out[7]: 'CN1C=CC=C1NC1=NC=CC=C1'
>
> Cheers,
> p.
>
> On 3 Jan 2020, at 23:26, Devendra Dhaked <dev...@gm...> wrote:
>
> 
> Hi,
>
> I have written SMIRKS for reaction transformation of  reagent
> "CN1C(NC2=NC=CC=C2)=CC=C1" into product "CN1CC=CC1=NC2=NC=CC=C2".
>
> but after transformation in forward direction it showed lost bond order in
> product molecule Cn1cC=C[c]1=Nc1ccccn1.
> code for forward direction
> reactant1 = Chem.MolFromSmiles('CN1C(NC2=NC=CC=C2)=CC=C1')
> rxn =
> AllChem.ReactionFromSmarts('[O,NX3;H1:2][cX3z2;r5:3]:[c;r5:4][c;r5:5]:[cr5R{1-2}:6]>>[O,S,NX2:2]=[CX3;z2;r5:3][C;r5:4]=[C;r5:5][CX4;r5;!H0:6]')
> ps=rxn.RunReactants((reactant1,))
> print(len(ps))
> print(Chem.MolToSmiles(ps[0][0],True))
>
> Similarly in backward direction produced product(reactant) also lost its
> bond order
> CN1cccc1Nc1ccccn1.
> code for backward direction
> reactant1 = Chem.MolFromSmiles('CN1CC=CC1=NC2=NC=CC=C2')
> rxn =
> AllChem.ReactionFromSmarts('[O,S,NX2:2]=[CX3;z2;r5:3][C;r5:4]=[C;r5:5][CX4;r5;!H0:6]>>[O,NX3;H1:2][cX3z2;r5:3]:[c;r5:4][c;r5:5]:[cr5R{1-2}:6]')
> ps=rxn.RunReactants((reactant1,))
> print(len(ps))
> print(Chem.MolToSmiles(ps[0][0],True))
>
> Thanks.
>
> --
> Regards
>
> Devendra K Dhaked (Ph.D.)
> Center for Cancer Research
> Chemical Biology Laboratory
> National Cancer Institute (NIH)
> Boyles Street, Frederick, MD
> USA
> _______________________________________________
> Rdkit-discuss mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>
>

-- 
Regards

Devendra K Dhaked (Ph.D.)
Center for Cancer Research
Chemical Biology Laboratory
National Cancer Institute (NIH)
Boyles Street, Frederick, MD
USA