|
3
|
1 OSRA: Optical Structure Recognition Application
|
|
|
2
|
|
|
3 OSRA is a utility designed to convert graphical representations of chemical
|
|
|
4 structures, as they appear in journal articles, patent documents, textbooks,
|
|
|
5 trade magazines etc., into SMILES (Simplified Molecular Input Line Entry
|
|
|
6 Specification - see http://en.wikipedia.org/wiki/SMILES) or
|
|
|
7 SD files - a computer recognizable molecular structure format.
|
|
|
8 OSRA can read a document in any of the over 90 graphical formats parseable by
|
|
|
9 ImageMagick - including GIF, JPEG, PNG, TIFF, PDF, PS etc., and generate
|
|
|
10 the SMILES or SDF representation of the molecular structure images encountered
|
|
|
11 within that document.
|
|
|
12
|
|
|
13 Note that any software designed for optical recognition is unlikely to be
|
|
|
14 perfect, and the output produced might, and probably will, contain errors,
|
|
|
15 so curation by a human knowledgeable in chemical structures is highly recommended.
|
|
|
16
|
|
|
17 http://cactus.nci.nih.gov/osra/
|
|
|
18
|
|
|
19 The wrapper comes with an automatic installation of all dependencies through the
|
|
|
20 galaxy toolshed.
|
