We have looked a number of times over the years at various sources of
reaction data that we might make available as part of eMolecules. We've
never found anything useful.
The entire patent system is wide open, but extracting meaningful reactions
from it is almost impossible due to the deliberate vagueness that patent
attorneys encourage. Worse, attempts to automatically extract reactions
from patents almost always turn up mostly solvents, chemicals that are
mentioned in passing and so forth. As I recall, several companies have
tried to extract reactions from patent literature, some even spending large
sums of money, and the results haven't met expectations.
It's hard work to extract reactions from literature. It can't be done
automatically; every reaction has to be reviewed and edited by an expert
chemist. The people who have done it in the past and are still doing it
want to be paid for it.
I can imagine setting up an open-reaction database, sort of like Wikipedia
for reactions, but the same problems that plague Wikipedia would have to be
addressed. You need some way to ensure quality. You might have
"gatekeepers" or registrars who have to approve every change. Or you could
have some sort of reputation system, with editing history, rollback, voting
and all sorts of other nonsense. It's a hard problem.
Craig
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk
_______________________________________________
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
