MINT and IntAct contribute to the Second BioCreative challenge: serving the text-mining community with high quality molecular interaction data
2008

MINT and IntAct Help Improve Protein Interaction Data

Sample size: 375 publication Evidence: moderate

Author Information

Author(s): Chatr-aryamontri Andrew, Kerrien Samuel, Khadake Jyoti, Orchard Sandra, Ceol Arnaud, Licata Luana, Castagnoli Luisa, Costa Stefano, Derow Cathy, Huntley Rachael, Aranda Bruno, Leroy Catherine, Thorneycroft Dave, Apweiler Rolf, Cesareni Gianni

Primary Institution: Department of Biology, University of Rome, Tor Vergata

Hypothesis

Can text-mining tools enhance the efficiency of curating protein-protein interaction data from literature?

Conclusion

Text-mining tools can improve the coverage of literature on protein-protein interactions by assisting manual curation.

Supporting Evidence

  • The study provided training and test datasets for evaluating text-mining tools.
  • Manual curation was found to be time-consuming and often incomplete.
  • Text-mining tools could help identify relevant sentences in full-text articles.

Takeaway

This study shows that using computers to help find information about how proteins interact can make it easier for scientists to gather data from research papers.

Methodology

The study involved manual curation of protein-protein interactions from literature and the development of text-mining tools to assist in this process.

Potential Biases

Discrepancies in curation between databases may lead to inconsistent data representation.

Limitations

The study faced challenges such as ambiguity in gene names and the inability to extract interactions solely from abstracts.

Digital Object Identifier (DOI)

10.1186/gb-2008-9-s2-s5

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication