I'm looking forward to these fixes. The current handling is wrong in many cases. I see no problem with replacing that with some version of "right" before 3.0. My current approach is to remove problematic molecules from my datasets.
- Noel 2009/10/29 Craig A. James <cja...@emolecules.com>: > Geoff and I have had a short private conversation about aromaticity, and I > had some ideas that are better expressed here on the discussion board. > > Currently, OpenBabel's aromaticity (typer.cpp) is pretty well broken. It > seems to accept most correct structures, ordinary stuff like benzenes and > pyridines, which is good. But it also accepts totally bogus structures like > this: > > echo "c1=c=c=c=c=c1" | babel -i smi -o smi > c1ccccc1 > 1 molecule converted > > This is so wrong it makes my head spin. > > It turns out the problem is fairly deep, and there is no simple solution to > it, and part of it goes even to the SMARTS parser itself. > > For my purposes, I just need it to work correctly for a cheminformatics > system, i.e. for SMILES and SDF formats. The SMILES definition of > aromaticity is clear, and the chemical concept of aromaticity is irrelevant > to me. > > For the future, the work that Greg is doing with RDKit and OpenBabel 3, which > will provide aromaticity plug-ins, is the right approach. > > Until that's available, I need to rewrite the aromaticity code in typer.cpp > and the rules in aromaticity.txt, to make them implement the SMILES > aromaticity rules correctly. But because the code will probably behave quite > differently than it does now, I don't think I should check it back in to the > Subversion repository. I'll just keep my own private version for use in our > cheminformatics system, and it (or a derivative of it) will only go in to > OpenBabel when version 3 plug-in technology is available. > > On the other hand, a different approach is just to make OpenBabel2 use a > strict SMILES aromaticity definition. I have trouble believing the current > aromaticity definition is good science; at least with a strict SMILES > definition, you'd know what you were getting. But then, I'm a mathmatical, > computer science type guy who took freshman chemistry and physics in college, > then hung around chemists for a few decades. I'm not a real chemist, so I'll > leave that judgement to others. > > Part 2 to follow... > > Craig > > ------------------------------------------------------------------------------ > Come build with us! The BlackBerry(R) Developer Conference in SF, CA > is the only developer event you need to attend this year. Jumpstart your > developing skills, take BlackBerry mobile applications to market and stay > ahead of the curve. Join us from November 9 - 12, 2009. Register now! > http://p.sf.net/sfu/devconference > _______________________________________________ > OpenBabel-Devel mailing list > OpenBabel-Devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/openbabel-devel > ------------------------------------------------------------------------------ Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9 - 12, 2009. Register now! http://p.sf.net/sfu/devconference _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
