You could also look into ProDy, if you use PDB files.
http://www.csb.pitt.edu/prody/reference/atomic/flags.html#term-hetero
----
Pozdrawiam, | Best regards,
Maciek Wójcikowski
[email protected]
2013/7/8 Maciek Wójcikowski <[email protected]>
> I think you need custom code. For python, you could use pybel. There are
> three lines that do what you want (
> http://openbabel.org/docs/dev/UseTheLibrary/Python_PybelAPI.html#pybel.Smarts.findall).
> You only miss one or two lines to get the atoms by their indices.
>
> Although searching pdb files with it could be a demanding and time
> consuming task. It's really made for small molecules, i guess. I'd
> personally would loop through atoms check if they are HETATOMs and then
> check SMARTS pattern. Alternatively, you can filter them by name, if you
> can suspect what's your residues name could be (although it's the most
> error prone).
>
> ----
> Pozdrawiam, | Best regards,
> Maciek Wójcikowski
> [email protected]
>
>
> 2013/7/8 Brett Hannigan <[email protected]>
>
>> Unfortunately, I don't think that's quite what I need. The obabel filter
>> command will look at the given pdb and then output the entire pdb if it
>> matches the given SMARTS query, right? What I would like to do is only get
>> the atoms that match the SMARTS query.
>>
>> For example, let's say I have a pdb with a single phospho-serine residue.
>> What I would like to do is get only the phosphate atoms from the pdb, and
>> not get the rest of the residue atoms.
>>
>> Really what I am going for here is I'd like specify some groups that I
>> find interesting, say a phosphate group, a phenyl group, etc. And then I
>> have a database of protein + ligand pdbs. I'd like to loop through those
>> pdbs and look for my groups that I am interested in. Rather than just
>> identify the pdbs that have the sub-structure I am interested in (i.e.
>> identify all pdbs with phosphates), I want to know which atoms in the pdb
>> constitute the interesting sub-structure. That way I could then align all
>> phosphate groups and look at the residues surrounding the phophates to see
>> if I see common modes of binding etc. Phosphate would just be one example,
>> you can imagine identifying many other interesting sub-structures.
>>
>> -Brett
>>
>> ------------------------------
>> *From:* Maciek Wójcikowski <[email protected]>
>> *To:* Brett T. Hannigan <[email protected]>
>> *Cc:* [email protected]
>> *Sent:* Saturday, July 6, 2013 6:21 AM
>> *Subject:* Re: [Open Babel] SMILES, SMARTS, and PDB
>>
>> Hello,
>>
>> You can "convert" from PDB to PDB using filter option (
>> http://openbabel.org/wiki/--filter_option)
>> obabel input.pdb -O output.pdb --filter "s='CN'"
>>
>> If your database is really large, or you want to search multiple times,
>> then I'd suggest using FastSearch for performance reasons. For further
>> lecture: http://openbabel.org/wiki/FastSearch
>>
>> ----
>> Pozdrawiam, | Best regards,
>> Maciek Wójcikowski
>> [email protected]
>>
>>
>> 2013/7/5 Brett T. Hannigan <[email protected]>
>>
>> I have a database of molecules in pdb files. I'd like to go through each
>> molecule and see if it has certain substructures, and if they do identify
>> which atoms are part of that substructure. So it's easy enough to read the
>> pdb file in, convert it to SMILES representation, and then do a SMARTS
>> query to look for my substructures. However, if the SMARTS query reports
>> that the substructure is present in the molecule, I'm not sure how to
>> identify the atoms in the pdb file which make up the match. Any ideas?
>>
>> --
>> ******************************************
>> Brett T. Hannigan
>> 8011 Ardleigh St.
>> Philadelphia, PA 19118
>> [email protected]
>> ******************************************
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net <http://sf.net/> email is sponsored by Windows:
>>
>> Build for Windows Store.
>>
>> http://p.sf.net/sfu/windows-dev2dev
>> _______________________________________________
>> OpenBabel-discuss mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>>
>>
>>
>>
>>
>
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:
Build for Windows Store.
http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
OpenBabel-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
