Fig. 2

Diagram of the annotation process. We collect publications in PubMed involving the HT methodologies. Searches were made for the term “coli” in the title or in the abstract and the name of the method or different synonyms or keywords related to the method in all fields of publications. The results were filtered to get unique results. We read the abstracts and eliminate all those papers not reporting experiments performed in E. coli. Frequently, the papers include additional experimental characterization for a subset of the sites based on classic methods. Metadata are extracted from each publication. For more detail see main text. The growth condition and strain are described using the controlled vocabulary defined by Tierrafría et al. [7]. The products of curation are added to RegulonDB either together with the classic curation or as a separate dataset. Image from RegulonDB [8]