Abstract
The exponential growth of research publications provides challenges for curators and researchers in finding and assimilating scientific facts described in the literature. Therefore, services that sup-port the browsing of articles and the identification of key concepts with minimal effort would be beneficial for the scientific community. Reference databases store such high value scientific facts and key concepts, in the form of annotations. Annotations are statements assigned by curators from an evidence in a publication. Yet, if annotated statements are linked with the publication’s references (e.g. PubMed identifiers), the evidences are rarely stored during the curation process. In this paper, we investigate the automatic relocalization of biological evidences, the Gene References Into Function (GeneRIFs), in scientific articles. GeneRIFs are free text statements extracted from an article, and potentially reformulated by a curator. De facto, only 33% of geneRIFs are copy-paste that can be retrieved by the reader with the search tool of his reader. For automatically retrieving the other evidences, we use an approximate string matching algorithm, based on a finite state automaton and a derivative Levenshtein distance. For evaluation, two hundred candidate sentences were evaluated by human experts. We present and compare results for the relocalization in both abstracts and fulltexts. With the optimal setting, 76% of the evidences are retrieved with a precision of 97%. This data free approach does not require any training data nor a priori lexical knowledge. Yet it remarkable how it handles with complex language modifications such as reformulations, acronyms expansion, or anaphora. In the whole MEDLINE, 350,000 geneRIFs were retrieved in abstracts, and 15,000 in fulltexts; they are currently available for highlighting in the Europe PMC literature browser.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Venkatesan, A., et al.: SciLite: a platform for displaying text-mined annotations as a means to link research articles with biological data. Wellcome Open Res. 1, 25 (2016). https://doi.org/10.12688/wellcomeopenres.10210.1
Howe, D., et al.: Big data: the future of biocuration. Nature 455(7209), 47–50 (2008). https://doi.org/10.1038/455047a
Gobeill, J., Pasche, E., Vishnyakova, D., Ruch, P.: Managing the data deluge: data-driven GO category assignment improves while complexity of functional annotation increases. Database (Oxford) (2013). https://doi.org/10.1093/database/bat041
Brown, G.R., et al.: Gene: a gene-centered information resource at NCBI. Nucl. Acids Res. 43(D1), D36–D42 (2015). https://doi.org/10.1093/nar/gku1055
Bultet, L.A., Aguilar-Rodriguez, J., Ahrens, C.H., Ahrne, E.L., Ai, N., et al.: The SIB Swiss Institute of Bioinformatics’ resources: focus on curated databases. Nucl. Acids Res. 44, D27–D37 (2016). https://doi.org/10.1093/nar/gkv1310
Baumgartner, W.A., Cohen, K.B., Fox, L.M., Acquaah-Mensah, G., Hunter, L.: Manual curation is not sufficient for annotation of genomic databases. Bioinformatics 23(13), i41–i48 (2007). https://doi.org/10.1093/bioinformatics/btm229
Jelier, R., et al.: Searching for geneRIFs: concept-based query expansion and Bayes classification. In: TREC Proceedings, pp. 225–233 (2003)
Obermeyer, Z., Emanuel, E.J.: Predicting the future - big data, machine learning, and clinical medicine. New Engl. J. Med. 375(13), 1216 (2016). https://doi.org/10.1056/NEJMp1606181
Tsuruoka, Y., Tsujii, J.I.: Improving the performance of dictionary-based approaches in protein name recognition. J. Biomed. Inform. 37(6), 461–470 (2004)
Papamichail, D., Papamichail, G.: Improved algorithms for approximate string matching. BMC Bioinform. 10(1), S10 (2009)
Wang, W., Xiao, C., Lin, X., Zhang, C.: Efficient approximate entity extraction with edit distance constraints. In: Proceedings of the 2009 ACM SIGMOD International Conference on Management of Data, pp. 759–770 (2009)
Buschmann, T., Bystrykh, L.V.: Levenshtein error-correcting barcodes for multiplexed DNA sequencing. BMC Bioinform. 14(1), 272 (2013)
Lasko, T.A., Hauser, S.E.: Approximate string matching algorithms for limited-vocabulary OCR output correction. In: Photonics West 2001-Electronic Imaging, pp. 232–240 (2000)
Wang, J., et al.: Interactive and fuzzy search: a dynamic way to explore MEDLINE. Bioinformatics 26(18), 2321–2327 (2010)
Hersh, W.R., Bhupatiraju, R.T.: TREC genomics track overview. In: TREC Proceedings, pp. 14–23 (2003)
Bhalotia, G., Nakov, P., Schwartz, A.S., Hearst, M.A.: BioText Team report for the TREC 2003 Genomics Track. In: TREC Proceedings, pp. 612–621 (2003)
Jimeno-Yepes, A.J., Sticco, J.C., Mork, J.G., Aronson, A.R.: GeneRIF indexing: sentence selection based on machine learning. BMC Bioinform. 14(1), 171 (2013)
Gobeill, J., Ruch, P., Zhou, X.: Query and document expansion with medical subject headings terms at medical Imageclef 2008. In: Peters, C., et al. (eds.) CLEF 2008. LNCS, vol. 5706, pp. 736–743. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-04447-2_95
Gobeill, J., et al.: Deep Question Answering for protein annotation. Database (Oxford) (2015). https://doi.org/10.1093/database/bav081
Pasche, E., Teodoro, D., Gobeill, J., Ruch, P., Lovis, C.: QA-driven guidelines generation for bacteriotherapy. In: AMIA Annual Symposium Proceedings, pp. 509–513 (2009)
Mottin, L., et al.: neXtA5: accelerating annotation of articles via automated approaches in neXtProt. Database 2016, baw098 (2016)
Levenshtein, V.I.: Binary codes capable of correcting deletions, insertions, and reversals. In: Soviet Physics Doklady, vol. 10, no. 8, pp. 707–710 (1966)
Wagner, R.A., Fischer, M.J.: The string-to-string correction problem. J. ACM (JACM) 21(1), 168–173 (1974)
Pustejovsky, J., Castano, J., Cochran, B., Kotecki, M., Morrell, M.: Automatic extraction of acronym-meaning pairs from MEDLINE databases. Stud. Health Technol. Inform. 1, 371–375 (2001)
Europe PMC Consortium: Europe PMC: a full-text literature database for the life sciences and platform for innovation. Nucl. Acids Res. (2014). https://doi.org/10.1093/nar/gku1061
Acknowledgments
This research was supported by the Elixir Excelerate project, funded by the European Commission within the Research Infrastructures programme of Horizon 2020, grant agreement number 676559. The authors thank their colleagues from the SIB Swiss Institute of Bioinformatics (Core-IT), in particular Daniel Texeira and Heinz Stockinger, who provided insight and expertise that greatly assisted the research. The authors also thank the European Bioinformatics Institute, in particular Johanna McEntyre and Aravind Venkatesan, for the integration of retrieved geneRIFs into EuropePMC.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 Springer Nature Switzerland AG
About this paper
Cite this paper
Gobeill, J., Pasche, E., Ruch, P. (2023). Retrieving the Evidence of a Free Text Annotation in a Scientific Article: A Data Free Approach. In: Gelbukh, A. (eds) Computational Linguistics and Intelligent Text Processing. CICLing 2019. Lecture Notes in Computer Science, vol 13451. Springer, Cham. https://doi.org/10.1007/978-3-031-24337-0_17
Download citation
DOI: https://doi.org/10.1007/978-3-031-24337-0_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-24336-3
Online ISBN: 978-3-031-24337-0
eBook Packages: Computer ScienceComputer Science (R0)