The Department of Computer Science and A.I. at the University of Granada together with the Center for Biomedical Informatics (CBI) at Harvard Medical School (HMS) are developing advance natural language processing (NLP) techniques to extract facts from biomedical literature. Particularly, our research concerns the automatic identification of gene-gene (protein-protein) interactions and gene-disease interactions.
To this end we have created BioNotate: a text annotation tool written in Javascript that allows collaborative annotation of relationships between biomedical entities in scientific texts.
- Anntotation Guidelines.
- Check out BioNotate and contribute to the current annotation effort here.
- Source code and corpora. Please check out the Download Section and Project Documentation / Installation Guide at the BioNotate Sourceforge.net project. The software is distributed under GPL.
A full description of the tool, the annotation schema and the results of a pilot annotation effort on autism disease network can be found in the paper:
Cano C., Monaghan T., Blanco A., Wall D.P. and Peshkin L.
"Collaborative text-annotation resource for disease-centered relation extraction from biomedical text"
Journal of Biomedical
Informatics, Volume 42, Issue 5, October 2009, Pages 967-977. PMID 19232400
