BioModels Database: a Unified resource for Systems Biology models

Even the simplest living organisms perform a huge number of different processes, which are interconnected in complex ways to ensure that the organism responds appropriately to its environment. One of the ways of ensuring that we really understand how these processes fit together is to build quantitative models of them, that can be simulated using computer. If a computer simulation behaves differently than the real organism, we know that we've neglected an important component of the system. Quantitative models can also reveal previously unappreciated properties of complex systems, for instance paving the way toward new drug treatments. This approach, known as 'Computational Systems Biology', is becoming increasingly popular now that scientists are accumulating detailed parts lists for many organisms, thanks to the 'omics' efforts to comprehensively document the components of living entities. BioModels Database provides access to quantitative models of biochemical and cell biological systems that have been published in the scientific literature, and verified to be accurate. Some of these models are very simple, containing just a few processes or reactions; others contain hundreds. The models are checked to verify that they behave as described in the reference publication. Human curators annotate and cross link components of the models to other relevant data resources. This allows users to identify precisely the components of models, and helps them to retrieve appropriate models, which they can then visualise and simulate using appropriate software. For instance, in order to develop a quantitative model of cell tumorigenesis, one may choose a suitable model of cell-cycle, and attempt to merge it with models of relevant cell signalling pathways such as the MAP kinase cascade. Only a database of trusted, peer-reviewed, annotated models will provide the adequate choice of 'building bricks'. Users can search models on many different criteria, visualise and and download them in various standard formats for further use. Over the last two years, BioModels Database has undergone an exponential growth. It is now the reference data resource for quantitative models. Deposition of models upon publication is advised by several scientific journals, and the atabase has been voted most used data resource in Systems Biology by an independent community survey. Developed as a proof of concept, BioModels Database has demonstrated its relevance and needs to evolve in a more mature resource in order to best serve the need of a growing UK Systems Biology community.

BioModels Database is a free resource of annotated, quantitative models of biological interest. The resource is a set of relational and XML-native databases, built around a pipeline that allows submission, curation, annotation and distribution of the models. Access to the database is handled by an application using JSP. Models submitted by authors, publishers, or coded from the literature, are verified and extensively annotated and linked to other data resources. They are then distributed in several formats, such as SBML, CellML, BioPAX, SciLab, XPP. The resource is made up of two different branches. The curated one distributes chemical kinetics models that can be entirely captured in SBML and quantitatively verified. The non-curated branch stores models that were not yet curated, models that could-not be properly verified, and models that are not currently entirely supported by SBML. As SBML evolved, more models will be incorporated in the curated branch. We are constantly improving and extending all parts of the data resource. Using new technologies and making better use of third-party information, we enhance the robustness and the efficiency of the core data resource, including data retrieval. We modify the curation, annotation and distribution interfaces, in order to offer a sustainable infrastructure able to support the development and maintainance of a comprehensive database of models. The pace of database population is increased by in-house curation and annotation, but also by strenghening our links with the publishing industry. An application programming interface and the corresponding WebServices shall allow third-parties to develop software with programmatic access to BioModels Database. To complement the file exports, we offer online simulation capabilities directly and through collaborations. Users will be able to pick models of interest, instantiate simulations using different simulation tools, and perform a first crude analysis of the results.