tmBioC: improving interoperability of text-mining tools with BioC
Authors:
Open Access
- 1 January 2014
- journal article
- research article
- Published by Oxford University Press (OUP) in Database: The Journal of Biological Databases and Curation
- Vol. 2014
Abstract
The lack of interoperability among biomedical text-mining tools is a major bottleneck in creating more complex applications. Despite the availability of numerous methods and techniques for various text-mining tasks, combining different tools requires substantial efforts and time owing to heterogeneity and variety in data formats. In response, BioC is a recent proposal that offers a minimalistic approach to tool interoperability by stipulating minimal changes to existing tools and applications. BioC is a family of XML formats that define how to present text documents and annotations, and also provides easy-to-use functions to read/write documents in the BioC format. In this study, we introduce our text-mining toolkit, which is designed to perform several challenging and significant tasks in the biomedical domain, and repackage the toolkit into BioC to enhance its interoperability. Our toolkit consists of six state-of-the-art tools for named-entity recognition, normalization and annotation (PubTator) of genes (GenNorm), diseases (DNorm), mutations (tmVar), species (SR4GN) and chemicals (tmChem). Although developed within the same group, each tool is designed to process input articles and output annotations in a different format. We modify these tools and enable them to read/write data in the proposed BioC format. We find that, using the BioC family of formats and functions, only minimal changes were required to build the newer versions of the tools. The resulting BioC wrapped toolkit, which we have named tmBioC, consists of our tools in BioC, an annotated full-text corpus in BioC, and a format detection and conversion tool. Furthermore, through participation in the 2013 BioCreative IV Interoperability Track, we empirically demonstrate that the tools in tmBioC can be more efficiently integrated with each other as well as with external tools: Our experimental results show that using BioC reduces >60% in lines of code for text-mining tool integration. The tmBioC toolkit is publicly available at http://www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/tmTools/. Database URL:http://www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/tmTools/Keywords
This publication has 24 references indexed in Scilit:
- DNorm: disease name normalization with pairwise learning to rankBioinformatics, 2013
- PubTator: a web-based text mining tool for assisting biocurationNucleic Acids Research, 2013
- BioC: a minimalist approach to interoperability for biomedical text processingDatabase: The Journal of Biological Databases and Curation, 2013
- An overview of the BioCreative 2012 Workshop Track III: interactive text mining taskDatabase: The Journal of Biological Databases and Curation, 2013
- Accelerating literature curation with text-mining tools: a case study of using PubTator to curate genes in PubMed abstractsDatabase: The Journal of Biological Databases and Curation, 2012
- A context-blocks model for identifying clinical relationships in patient recordsBMC Bioinformatics, 2011
- Semi-automatic semantic annotation of PubMed queries: A study on quality, efficiency, satisfactionJournal of Biomedical Informatics, 2011
- Extracting Rx information from clinical narrativeJournal of the American Medical Informatics Association, 2010
- U-Compare: share and compare text mining tools with UIMABioinformatics, 2009
- Understanding PubMed(R) user search behavior through log analysisDatabase: The Journal of Biological Databases and Curation, 2009