SUMMARY: RegulonDB has collected, harmonized and centralized data from hundreds of experiments for nearly two decades and is considered a point of reference for transcriptional regulation in Escherichia coli K12. Here, we present the regutools R package to facilitate programmatic access to RegulonDB data in computational biology. regutools gives researchers the possibility of writing reproducible workflows with automated queries to RegulonDB. The regutools package serves as a bridge between RegulonDB data and the Bioconductor ecosystem by reusing the data structures and statistical methods powered by other Bioconductor packages. We demonstrate the integration of regutools with Bioconductor by analyzing transcription factor DNA binding sites and transcriptional regulatory networks from RegulonDB. We anticipate that regutools will serve as a useful building block in our progress to further our understanding of gene regulatory networks. AVAILABILITY AND IMPLEMENTATION: regutools is an R package available through Bioconductor at bioconductor.org/packages/regutools.

MOTIVATION: A major component in increasing our understanding of the biology of an organism is the mapping of its genotypic potential into its phenotypic expression profiles. This mapping is executed by the machinery of gene regulation, which is essentially studied by changes in growth conditions. Although many efforts have been made to systematize the annotation of experimental conditions in microbiology, the available annotations are not based on a consistent and controlled vocabulary, making difficult the identification of biologically meaningful comparisons of knowledge derived from different experiments or laboratories. RESULTS: We curated terms related to experimental conditions that affect gene expression in Escherichia coli K-12. Since this is the best-studied microorganism, the collected terms are the seed for the Microbial Conditions Ontology (MCO), a controlled and structured vocabulary that can be expanded to annotate microbial conditions in general. Moreover, we developed an annotation framework to describe experimental conditions, providing the foundation to identify regulatory networks that operate under particular conditions. AVAILABILITY AND IMPLEMENTATION: As far as we know, MCO is the first ontology for growth conditions of any bacterial organism, and it is available at http://regulondb.ccg.unam.mx and https://github.com/microbial-conditions-ontology. Furthermore, we will disseminate MCO throughout the Open Biological and Biomedical Ontology (OBO) Foundry in order to set a standard for the annotation of gene expression data. This will enable comparison of data from diverse data sources. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

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.

BACKGROUND: Escherichia coli is the model organism for which our knowledge of its regulatory network is the most extensive. Over the last few years, our project has been collecting and curating the literature concerning E. coli transcription initiation and operons, providing in both the RegulonDB and EcoCyc databases the largest electronically encoded network available. A paper published recently by Ma et al. (2004) showed several differences in the versions of the network present in these two databases. Discrepancies have been corrected, annotations from this and other groups (Shen-Orr et al., 2002) have been added, making the RegulonDB and EcoCyc databases the largest comprehensive and constantly curated regulatory network of E. coli K-12. RESULTS: Several groups have been using these curated data as part of their bioinformatics and systems biology projects, in combination with external data obtained from other sources, thus enlarging the dataset initially obtained from either RegulonDB or EcoCyc of the E. coli K12 regulatory network. We kindly obtained from the groups of Uri Alon and Hong-Wu Ma the interactions they have added to enrich their public versions of the E. coli regulatory network. These were used to search for original references and curate them with the same standards we use regularly, adding in several cases the original references (instead of reviews or missing references), as well as adding the corresponding experimental evidence codes. We also corrected all discrepancies in the two databases available as explained below. CONCLUSION: One hundred and fifty new interactions have been added to our databases as a result of this specific curation effort, in addition to those added as a result of our continuous curation work. RegulonDB gene names are now based on those of EcoCyc to avoid confusion due to gene names and synonyms, and the public releases of RegulonDB and EcoCyc are henceforth synchronized to avoid confusion due to different versions. Public flat files are available providing direct access to the regulatory network interactions thus avoiding errors due to differences in database modelling and representation. The regulatory network available in RegulonDB and EcoCyc is the most comprehensive and regularly updated electronically-encoded regulatory network of E. coli K-12.

MOTIVATION: As one of the best-characterized free-living organisms, Escherichia coli and its recently completed genomic sequence offer a special opportunity to exploit systematically the variety of regulatory data available in the literature in order to make a comprehensive set of regulatory predictions in the whole genome. RESULTS: The complete genome sequence of E.coli was analyzed for the binding of transcriptional regulators upstream of coding sequences. The biological information contained in RegulonDB (Huerta, A.M. et al., Nucleic Acids Res.,26,55-60, 1998) for 56 different transcriptional proteins was the support to implement a stringent strategy combining string search and weight matrices. We estimate that our search included representatives of 15-25% of the total number of regulatory binding proteins in E.coli. This search was performed on the set of 4288 putative regulatory regions, each 450 bp long. Within the regions with predicted sites, 89% are regulated by one protein and 81% involve only one site. These numbers are reasonably consistent with the distribution of experimental regulatory sites. Regulatory sites are found in 603 regions corresponding to 16% of operon regions and 10% of intra-operonic regions. Additional evidence gives stronger support to some of these predictions, including the position of the site, biological consistency with the function of the downstream gene, as well as genetic evidence for the regulatory interaction. The predictions described here were incorporated into the map presented in the paper describing the complete E.coli genome (Blattner,F.R. et al., Science, 277, 1453-1461, 1997). AVAILABILITY: The complete set of predictions in GenBank format is available at the url: http://www. cifn.unam.mx/Computational_Biology/E.coli-predictions CONTACT: [email protected], [email protected]

MOTIVATION: One of the most common methodologies to identify cis-regulatory sites in regulatory regions in the DNA is that of weight matrices, as testified by several articles in this issue. An alternative to strengthen the computational predictions in regulatory regions is to develop methods that incorporate more biological properties present in such DNA regions. The grammatical implementation presented in this paper provides a concrete example in this direction. RESULTS: On the basis of the analysis of an exhaustive collection of regulatory regions in Escherichia coli, a grammatical model for the regulatory regions of sigma 70 promoters has been developed. The terminal symbols of the grammar represent individual sites for the binding of activator and repressor proteins, and include the precise position of sites in relation to transcription initiation. Combining these symbols, the grammar generates a large number of different sentences, each of which can be searched for matching against a collection of regulatory regions by means of weight matrices specific for each set of sites for individual proteins. On the basis of this grammatical model, a Prolog syntactic recognizer is presented here. Specific subgrammars for ArgR, LexA and TyrR were implemented. When parsing a collection of 128 sigma 70 promoter regions, the syntactic recognizer produces a much lower number of false-positive sites than the standard search using weight matrices.