"Note that any software designed for optical recognition is unlikely to be perfect, and the output produced might, and probably will, contain errors, so a curation by a human knowledgeable in chemical structures is highly recommended." [OSRA]As found in Peter Corbett's bookmarking network has the National Cancer Insitute (NCI) released an Optical Structure Recognition Algorithm (OSRA) under the GPL license. Compared to the already available programs is this program open source, and even free of charge. The chemical part of the program uses OpenBabel.
As said, nobody expects that the program is perfect in transforming images to molecular structures. I am wondering if Wikinomics might help in curating such data?
References
- "This is not a pipe", Magritte
Posts
4 comments:
My molecular recognition system detects an unstable anti-aromatic (12-pi electron) structure in the figure.
4n+2 ... hahaha ;-)
As I have just realized have Egon and Antony also blogged about OSRA. Social networks are funny and work, but only if you have time checking what the others are doing ;-)
Joerg, as you comment...social communities only work when you check or the aggregators are having the right turnaround and dumping them into our readers...and we STILL have to check. Maybe I'll redirect to my blackberry somehow...that I check.
As a first release I say it's a good step in the right direction. I know that I'll be spending more time on this in the foreseeable future and I think it's a good contribution to the community. Imperfect for sure...but so am I...and I have my uses. And I certainly know a few people who wish they could redraw parts of me...
Post a Comment