Re: [Rdkit-discuss] Why this doesn't work? HasStructMatch function

[Li, X. [xiaoboli] <Xia...@li...>]

Hi Greg,

Thanks. Now I understand it.

Best regards,

Dr. Xiaobo Li

Postdoctoral Research Associate
Andrew I. Cooper Group

University of Liverpool / Materials Innovation Factory
Department of Chemistry
Crown Street
Liverpool, L69 7ZD
United Kingdom

From: Greg Landrum [mailto:gre...@gm...]
Sent: 20 February 2019 07:37
To: Li, Xiaobo [xiaoboli] <Xia...@li...>
Cc: Rdk...@li...
Subject: Re: [Rdkit-discuss] Why this doesn't work? HasStructMatch function

The middle ring in your system is considered to be aromatic. You can see this by looking at the output SMILES:

In [3]: m=Chem.MolFromSmiles('CN(C(C=CC=C1)=C1C2=O)C3=C2C=CC=C3')

In [4]: Chem.MolToSmiles(m)
Out[4]: 'Cn1c2ccccc2c(=O)c2ccccc21'

Given this, you need either the ~ bond (which matches anything) instead of -, or you could avoid specifying the bond order in your query. In SMARTS a bond that's not specified is either single or aromatic:

In [5]: s=Chem.MolFromSmarts('c1ccccc1[#6](c2ccccc2)=[#8]')

In [6]: m.HasSubstructMatch(s)
Out[6]: True

Best,
-greg


On Tue, Feb 19, 2019 at 11:56 PM Jason Biggs <jas...@gm...<mailto:jas...@gm...>> wrote:
Change your pattern to use ~ as an unspecified bond instead of - for a single bond:

>m=Chem.MolFromSmiles('CN(C(C=CC=C1)=C1C2=O)C3=C2C=CC=C3')
>s=Chem.MolFromSmarts('c1ccccc1~[#6](~c2ccccc2)=[#8]')
>m.HasSubstructMatch(s)

True


Jason


On Tue, Feb 19, 2019 at 4:42 PM Li, Xiaobo [xiaoboli] <Xia...@li...<mailto:Xia...@li...>> wrote:

Dear all,



Why the output is False?



m=Chem.MolFromSmiles('CN(C(C=CC=C1)=C1C2=O)C3=C2C=CC=C3')

s=Chem.MolFromSmarts('c1ccccc1-[#6](-c2ccccc2)=[#8]')

m.HasSubstructMatch(s)

Output: False

m

[cid:image001.png@01D4C93D.34643F00]



s

[cid:image002.png@01D4C93D.34643F00]





Best regards,



Xiaobo Li








_______________________________________________
Rdkit-discuss mailing list
Rdk...@li...<mailto:Rdk...@li...>
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
_______________________________________________
Rdkit-discuss mailing list
Rdk...@li...<mailto:Rdk...@li...>
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss