CTD

As part of our participation in BioCreative 2012 and BioCreative 2013 we have performed several experiments with the Comparative Toxicogenomics Database (CTD). We have used CTD datasets to train and test entity extraction and relation extraction algorithms (an example of extracted relations is shown in the picture to the right). Additionally, we have customized ODIN using CTD entities (see picture left). We have obtained the BEST overall results in the CTD task in BioCreative 2012.

This activity is described in the publications listed below.






Related publications

  1. Fabio Rinaldi, Simon Clematide, Simon Hafner, Gerold Schneider, Gintare Grigonyte, Martin Romacker, Therese Vachon. Using the OntoGene pipeline for the triage task of BioCreative 2012, The Journal of Biological Databases and Curation, Oxford Journals, 2012. doi:10.1093/database/bas053, PMC3568389, PMID: 23396322
  2. Fabio Rinaldi, Simon Clematide, Tilia Renate Ellendorff, Hernani Marques. OntoGene: CTD entity and action term recognition. Proceedings of the Fourth BioCreative Challenge Evaluation Workshop, vol. 1, pages 90-94. Bethesda, Maryland, October 7-9, 2013.
  3. Fabio Rinaldi, Allan Peter Davis, Christopher Southan, Simon Clematide, Tilia Renate Ellendorff, Gerold Schneider. ODIN: a customizable literature curation tool.   Proceedings of the Fourth BioCreative Challenge Evaluation Workshop, vol. 1, pages 219-223. Bethesda, Maryland, October 7-9, 2013.
  4. Tilia Ellendorff, Fabio Rinaldi, Simon Clematide. Using Large Biomedical Databases as Gold Annotations for Automatic Relation Extraction. Proceedings of LREC 2014: The ninth international conference on Language Resources and Evaluation. Reykjavik, Iceland. 
Comments