Linking entries in protein interaction database to structured text: The FEBS Letters experiment

The corpus of the scientific literature has reached such size that a lot of useful data, dispersed throughout millions different articles, are now hard to recover. For instance, many articles in the biological domain describe relationships between entities (gene, proteins, small molecules, etc.) yet this crucial information cannot be efficiently used because of the difficulties in retrieving it automatically from unstructured text. Databases are striving to capture this valuable information and to organize it in a structured format ready for automatic analysis. However, the current database model, based on manual curation, is not sustainable because the limited support is not compatible with complete and accurate coverage of published information. Several proposals have been put forward to increase the efficiency and accuracy of the curation process. Here we present an experiment, designed by the editorial board of FEBS Letters, aimed at integrating each manuscript with a structured summary precisely reporting, with database identifiers and predefined controlled vocabularies, the protein interactions reported in the manuscript. The authors play an important role in this process as they are requested to provide structured information to be appended, in the form of human-readable paragraphs, at the end of traditional summaries. It is envisaged that the structured text will become an integral part of Medline abstracts. In 6 months time the experience gained with this experiment will form the basis for a community discussion to propose a widely accepted strategy for information storage and retrieval.