David,

 Thank you very much for the reply, very smart hack!

Sam

On Wed, Aug 31, 2016 at 7:27 PM, David Hall <li...@cowsandmilk.net> wrote:

> It is a bit of a hack, but for certain applications where I know it is
> likely fine, I use networkx to do subgraph isomorphism while considering
> only the elements and not looking at the order of the bonds. I can
> trivially draw a molecule where this does the wrong thing, but the majority
> of the time, it does the right thing.
>
> For the system you are discussing, here's an example using the attached
> script:
>
> $ ./isomorph.py 1PE_model.sdf 1kf6_1pe.pdb
> [{0: 0,
>   1: 1,
>   2: 2,
>   3: 3,
>   4: 4,
>   5: 5,
>   6: 6,
>   7: 7,
>   8: 8,
>   9: 9,
>   10: 10,
>   11: 11,
>   12: 12,
>   13: 13,
>   14: 14,
>   15: 15},
>  {0: 15,
>   1: 13,
>   2: 14,
>   3: 12,
>   4: 11,
>   5: 10,
>   6: 9,
>   7: 8,
>   8: 7,
>   9: 6,
>   10: 5,
>   11: 4,
>   12: 3,
>   13: 1,
>   14: 2,
>   15: 0}]
>
> Since 1PE is symmetric, it outputs the forward ordering of the 1st -> 1st,
> 2nd -> 2nd, etc. and reverse ordering of 1st->16th, 2nd  -> 15th.
>
> Hopefully the attached script gives an idea of how to incorporate this
> idea into your code.
>
>
> On Wed, Aug 31, 2016 at 4:51 PM, Sam Tonddast-Navaei <s.tondd...@gmail.com
> > wrote:
>
>> I am trying to read a small molecule from PDB file and match its atom
>> numbers to the same molecule in a SDF file. I have tried both matching
>> SMART patterns and using OBIsomorphismMapper, both work for 70% of
>> cases. However there are cases for which OpenBabel can not simply get the
>> right SMILES from just the PDB file (e.g.  adds extra double bonds or miss
>> ones). For example in PDB ID 1KF6 and ligand name IPE, if you generate
>> SMILES from PDB you will get OCCO/C=C/OCCO/C=C/O/C=C/O while the correct
>> one should be OCCOCCOCCOCCOCCO. I would appreciate if someone can give me
>> some advise on this.
>>
>> Thanks,
>> Sam
>>
>> ------------------------------------------------------------
>> ------------------
>>
>> _______________________________________________
>> OpenBabel-discuss mailing list
>> OpenBabel-discuss@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>>
>>
>
------------------------------------------------------------------------------
_______________________________________________
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss

Reply via email to