Ok - I understand now, and your comments are definitely taken on board.
It's certainly true that OBChemTsfm is missing many features of a SMIRKS
transformer, and it would be great to have more in there. I don't really
know how difficult that would be; I'll add it to my list, but I'd
definitely encourage anyone who's interested to see whether they could
improve on it. We're putting out calls for Google Summer of Code at the
moment; maybe this is a possible project for a student.
On 1 February 2018 at 18:27, Stefano Forli <fo...@scripps.edu> wrote:
> Thanks for the clarification.
>
> On 02/01/2018 12:43 AM, Noel O'Boyle wrote:
>
>> Actually, I think that this particular is most easily solved as Chris has
>> suggested, using regular ring closure digits. The deferred ring closures
>> (&) are currently not accepted by OB, and I guess aren't really needed in
>> any case. We should get them working though. They are a feature of SMILES
>> that is not very well documented.
>>
>> We have support for RXN, but the question is not about representing
>> reactions but generating molecules.
>>
>
> You are right, I was not clear on the matter.
> The RXN file format is supported, but not the information contained in it:
> in other words, you can't use a RXN file to instruct an OB class
> (OBChemTsfm?) to perform a chemical reaction.
>
> With that in place, I found that it would be extremely easy to generate
> large virtual libraries in this way:
> - tag fragment breaking points/attachments with unusual elements, i.e.,
> W, Re, Os, Ir, etc. I use each element to tag different kinds of fragments,
> so W can be aromatic rings, Re heterocycles, Os aliphatic chains, etc.
>
> - sketch a RXN file with a simple reaction such as: frag1-W + frag2-Re
> -> frag1-frag2
>
> In both Indigo and RDKit, loading and performing the reaction can be done
> in a few lines of Python, and for me it did the trick.
>
>
>> I have just written up a proposal to change how we handle reactions in
>> Open Babel. If/when this is in place, it would be worth revisiting support
>> for reaction transformations. In the meanwhile, if you are aware of any
>> problems with OBChemTsfm, we'd appreciate if you could file an issue.
>>
>
> I already discussed the issues in the mailing list a while back (2014,
> maybe?), but you suggested that without atom-mapping of reactions,
> OBChemTsfm does not work reliably.
>
> If it can be helpful to get an idea about the issue, my example was the
> following:
>
> import pybel
> ob = pybel.ob
> start = '[O:1]=[C:2]-[C:3]=[C:4]'
> end = '[O:1]=[C:2]-[C:3]-[C:4]-[S:5][C:6]'
> transformer = ob.OBChemTsfm()
> transformer.Init(start, end)
> org = 'C=CC(=O)Nc1ccccc1'
> mol = pybel.readstring('can', org)
> transformer.Apply(mol.OBMol)
> print "Original :", org
> print "Modified :", mol.write('can')
>
> Essentially, adding atoms to a molecule with OBChemTsfm is not possible.
> I've tried adding one atom at the time exploiting explicit hydrogens in the
> reaction, but that doesn't work either.
>
> I believe that mapping reactions would be possible once there would be
> some support for identifying MCSS, and I tried tackling the problem, back
> then. Although, the effort was beyond my time budget, so I had to give up
> (and use other toolkits, alas).
>
>
> S
>
>
> p.s. apologies for the repeated emails on the subject, but our spam filter
> was making a mess and I couldn't see my own posts to the list
>
>
>> Regards,
>> - Noel
>>
>>
>> On 1 February 2018 at 00:16, Stefano Forli <fo...@scripps.edu <mailto:
>> fo...@scripps.edu>> wrote:
>>
>> My experience with simulating reactions with OpenBabel was mostly
>> unsuccessful.
>> I've tried using OBChemTsfm to atoms, but I couldn't have it to work
>> even by adding
>> one atom at the time.
>>
>> The most straightforward way to deal with a library enumeration would
>> be to have
>> support for the RXN format, which can be used to describe any
>> arbitrary reaction. For
>> that, I've found pretty easy to work with Indigo and RDKit.
>>
>> On the other hand, I'm not familiar with the "&" notation Noel is
>> referring to.
>> If there's some documentation on the matter, please let me know!
>>
>> Thanks,
>>
>> S
>>
>> On 01/30/2018 11:59 AM, Nicolas Cheron wrote:
>>
>> Please allow me to do shameless self-promotion. I have developped
>> an open-source
>> program (using OpenBabel) that aims at creating de novo ligands
>> by directly
>> growing them in the active site of a target protein. You can also
>> use it without
>> the protein (I have called it FOG2) : you can define a core
>> molecule (so-called
>> seed), choose from which atom you will grow new molecules, and
>> the program will
>> create new molecules by adding fragments to the seed in a
>> "rational" way. If you
>> want more details, let me know, I can explain it.
>>
>> The links: http://pubs.acs.org/doi/10.1021/acs.jmedchem.5b00886
>> <http://pubs.acs.org/doi/10.1021/acs.jmedchem.5b00886>,
>> https://sourceforge.net/projects/opengrowth/
>> <https://sourceforge.net/projects/opengrowth/>
>>
>> Nicolas
>>
>>
>>
>> 2018-01-30 20:53 GMT+01:00 Chris Swain <sw...@mac.com <mailto:
>> sw...@mac.com>
>> <mailto:sw...@mac.com <mailto:sw...@mac.com>>>:
>>
>> Hi,
>>
>> It depends on the transformations you want to undertake. I
>> did a very quick
>> library
>> build about 20 years ago using SMILES strings. Just have
>> dummy atoms with unique
>> letters where you want to add functional groups then just do
>> a text find and
>> replace
>> with the desired SMILES of the building block you want to
>> add (which can of
>> course
>> contain dummy atoms). Then run through OpenBabel to generate
>> canonical SMILES. I
>> generated a 10 million molecule library doing this.
>>
>> Cheers,
>>
>> Chris
>>
>> On 30 Jan 2018, at 19:45, openbabel-discuss-request@list
>> s.sourceforge.net
>> <mailto:openbabel-discuss-requ...@lists.sourceforge.net>
>> <mailto:openbabel-discuss-requ...@lists.sourceforge.net
>> <mailto:openbabel-discuss-requ...@lists.sourceforge.net>>
>> wrote:
>>
>> Message: 1
>> Date: Tue, 30 Jan 2018 15:25:38 +0000 (UTC)
>> From: Siavoush Dastmalchi <siavous...@yahoo.com
>> <mailto:siavous...@yahoo.com> <mailto:siavous...@yahoo.com
>> <mailto:siavous...@yahoo.com>>>
>> To: "openbabel-disc...@lists.sf.net
>> <mailto:openbabel-disc...@lists.sf.net> <mailto:
>> openbabel-disc...@lists.sf.net
>> <mailto:openbabel-disc...@lists.sf.net>>"
>> <openbabel-disc...@lists.sf.net <mailto:
>> openbabel-disc...@lists.sf.net>
>> <mailto:openbabel-disc...@lists.sf.net <mailto:
>> openbabel-disc...@lists.sf.net>>>
>> Subject: [Open Babel] using Open Babel for virtual
>> library generation
>> Message-ID: <1885905137.1937075.1517325938
>> 7...@mail.yahoo.com
>> <mailto:1885905137.1937075.1517325938...@mail.yahoo.com>
>> <mailto:1885905137.1937075.1517325938...@mail.yahoo.com
>> <mailto:1885905137.1937075.1517325938...@mail.yahoo.com>>>
>> Content-Type: text/plain; charset="utf-8"
>>
>> Dear List,
>> I was wondering if it is possible to use OB for virtual
>> library
>> generation. For
>> example adding building blocks in a file to different
>> sites of a core
>> structure in a
>> combinatorial manner.
>> Many thanks in advance for you help.
>> Cheers,Siavoush
>> -------------- next part --------------
>> An HTML attachment was scrubbed...
>>
>>
>>
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's
>> most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> _______________________________________________
>> OpenBabel-discuss mailing list
>> OpenBabel-discuss@lists.sourceforge.net
>> <mailto:OpenBabel-discuss@lists.sourceforge.net>
>> <mailto:OpenBabel-discuss@lists.sourceforge.net
>> <mailto:OpenBabel-discuss@lists.sourceforge.net>>
>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>> <https://lists.sourceforge.net/lists/listinfo/openbabel-discuss>
>> <https://lists.sourceforge.net/lists/listinfo/openbabel-disc
>> uss
>> <https://lists.sourceforge.net/lists/listinfo/openbabel-discuss>>
>>
>>
>>
>>
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>
>>
>>
>> _______________________________________________
>> OpenBabel-discuss mailing list
>> OpenBabel-discuss@lists.sourceforge.net
>> <mailto:OpenBabel-discuss@lists.sourceforge.net>
>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>> <https://lists.sourceforge.net/lists/listinfo/openbabel-discuss>
>>
>>
>> --
>> Stefano Forli, PhD
>>
>> Assistant Professor
>> Dept. of Integrative Structural
>> and Computational Biology, MB-112A
>> The Scripps Research Institute
>> 10550 North Torrey Pines Road
>> La Jolla, CA 92037-1000, USA.
>>
>> tel: +1 (858)784-2055
>> fax: +1 (858)784-2860
>> email: fo...@scripps.edu <mailto:fo...@scripps.edu>
>> http://www.scripps.edu/~forli/
>>
>>
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> _______________________________________________
>> OpenBabel-discuss mailing list
>> OpenBabel-discuss@lists.sourceforge.net <mailto:
>> OpenBabel-discuss@lists.sourceforge.net>
>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>> <https://lists.sourceforge.net/lists/listinfo/openbabel-discuss>
>>
>>
>>
> --
>
> Stefano Forli, PhD
>
> Assistant Professor
> Dept. of Integrative Structural
> and Computational Biology, MB-112A
> The Scripps Research Institute
> 10550 North Torrey Pines Road
> La Jolla, CA 92037-1000, USA.
>
> tel: +1 (858)784-2055
> fax: +1 (858)784-2860
> email: fo...@scripps.edu
> http://www.scripps.edu/~forli/
>
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss