On 2013-11-16 10:21, David Hall wrote: ... > Of those strings, only "SMILES_CANONICAL CACTVS" specify two > stereochemical centers. The ones without canonical in their name don't > specify any stereochemistry at all. I'm pretty sure those never specify > any stereochemistry in the entire ligand expo. ... > Looking at the mmcif and experimental coordinates in pymol, the rcsb > completely screwed them up.
That doesn't surprise me. I should have generated the images and looked -- it was late on Friday is my excuse. ... Openbabel will not give you > stereochemistry when the stereochemistry is not specified in the input. > In these cases, you've provided, you say that CACTVS and OpenEye give > you certain stereochemistries in the inchi output or in general. In my > quick test: > > obabel -:"CCC(C)(C)C(=O)C(=O)N1CCCCC1C(=O)OC(CCc2ccccc2)c3ccccc3" -oinchi > ============================== > *** Open Babel Warning in InChI code > #1 :Omitted undefined stereo > InChI=1S/C28H35NO4/c1-4-28(2,3)25(30)26(31)29-20-12-11-17-23(29)27(32)33-24(22-15-9-6-10-16-22)19-18-21-13-7-5-8-14-21/h5-10,13-16,23-24H,4,11-12,17-20H2,1-3H3 > > openbabel does not give any stereochemistry in the output inchi. And it > warns you that there is undefined stereochemistry. Not sure why you are > claiming differently. Probably because I'm running a python script that after pybel.readstring/readfile() does mol.make3D() mol.addh() OBStereoFacade( mol, True ) which supposedly calls PerceiveStereo() which I'm not finding in the docs but presume it tries to figure out the stereochemistry. It flags 2 'R' carbons for 4 out of 5 SMILES strings that you say don't have stereochemistry. R's are consistent with 5th SMILES string that does specify stereochemistry. That's great, What's not great it generates different inchi strings for all of them because then I don't have a primary key for my database. The other not so great thing is that whatever perceives those R's doesn't work when the input is an inchi string. As for the warning, even if there is one it's very likely lost between swig and pybel. There's plenty of stuff swig doesn't do for you. (E.g. the latest weirdness I found was that "mol.addh() ; mol.make3D()" segfaults wheres "mol.make3D() ; mol.addh()" doesn't -- or maybe it's the other way around, I forget.) Cheers Dima ------------------------------------------------------------------------------ DreamFactory - Open Source REST & JSON Services for HTML5 & Native Apps OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access Free app hosting. Or install the open source package on any LAMP server. Sign up and see examples for AngularJS, jQuery, Sencha Touch and Native! http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk _______________________________________________ OpenBabel-discuss mailing list OpenBabel-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
