Automatic reconstruction of a bacterial regulatory network using Natural Language Processing

Abstract

Background: Manual curation of biological databases, an expensive and labor-intensive process, is essential for high quality integrated data. In this paper we report the implementation of a state-of-the-art Natural Language Processing system that creates computer-readable networks of regulatory interactions directly from different collections of abstracts and full-text papers. Our major aim is to understand how automatic annotation using Text-Mining techniques can complement manual curation of biological databases. We implemented a rule-based system to generate networks from different sets of documents dealing with regulation in Escherichia coli K-12.

Results: Performance evaluation is based on the most comprehensive transcriptional regulation database for any organism, the manually-curated RegulonDB, 45% of which we were able to recreate automatically. From our automated analysis we were also able to find some new interactions from papers not already curated, or that were missed in the manual filtering and review of the literature. We also put forward a novel Regulatory Interaction Markup Language better suited than SBML for simultaneously representing data of interest for biologists and text miners.

Conclusion: Manual curation of the output of automatic processing of text is a good way to complement a more detailed review of the literature, either for validating the results of what has been already annotated, or for discovering facts and information that might have been overlooked at the triage or curation stages.

Figure 1

Annotation workflow. A suggested workflow for parallel manual and automatic annotations of transcriptional regulation, with manual review of automatically-generated networks in shaded lines. Curators would check the interactions mined from text, since they would be provided with the reference papers and the textual segments from which the system retrieved them.

Figure 2

Corpus coverage of transcriptional regulation in E. Coli. Venn diagram illustrates overlapping coverage in corpora used in this work, with dots representing papers relevant for transcriptional regulation in E. coli K-12. Different selection strategies (keyword searches on PubMed and curated databases references) result in diverse document sets, which can contain in some cases groups of the same documents as well as other non-relevant papers.

Figure 3

Curated and retrieved articles for RegulonDB, by year. Comparison between all references initially retrieved from PubMed using RegulonDB curator's search algorithms, and references that were finally reviewed in full to populate the database. Since search algorithms are refined and changed continuously, this is shown only for illustration.