Where academic tradition
meets the exciting future

Complex-to-Pairwise Mapping of Biological Relationships Using a Semantic Network Representation

Juho Heimonen, Sampo Pyysalo, Filip Ginter, Tapio Salakoski, Complex-to-Pairwise Mapping of Biological Relationships Using a Semantic Network Representation. In: Tapio Salakoski, Dietrich Rebholz-Schuhmann, Sampo Pyysalo (Eds.), Proceedings of the Third International Symposium on Semantic Mining in Biomedicine (SMBM 2008), 45–52, Turku Centre for Computer Science (TUCS), 2008.

Abstract:

This study examines representations of protein-protein interactions focusing on the mapping between simple, pairwise annotation and complex, structured annotation. A simple semantic network representation equivalent to the BioInfer predicate formalism is introduced and used to transform the complex annotation of BioInfer into pairwise annotation through hand-written rules. Evaluation shows that this binarisation can be largely validly performed with limited loss of information, but also reveals specific challenges. The binarised BioInfer is the first corpus of this type where the inclusion rules are formalised to the level of a computational implementation and is freely available at http://www.it.utu.fi/BioInfer.

Files:

Full publication in PDF-format

BibTeX entry:

@INPROCEEDINGS{inpHePyGiSa08a,
  title = {Complex-to-Pairwise Mapping of Biological Relationships Using a Semantic Network Representation},
  booktitle = {Proceedings of the Third International Symposium on Semantic Mining in Biomedicine (SMBM 2008)},
  author = {Heimonen, Juho and Pyysalo, Sampo and Ginter, Filip and Salakoski, Tapio},
  editor = {Salakoski, Tapio and Rebholz-Schuhmann, Dietrich and Pyysalo, Sampo},
  publisher = {Turku Centre for Computer Science (TUCS)},
  pages = {45–52},
  year = {2008},
  keywords = {information extraction, protein-protein interactions, semantic networks},
}

Belongs to TUCS Research Unit(s): Turku BioNLP Group

Edit publication