YEASTNET: A consensus reconstruction of yeast metabolism

Latest version

Yeast consensus reconstructions are now made available at, this page is left here only for historical purposes

Version 3

Version 3 of this reconstruction was released in November 2009, and consists of 2830 reactions (including 2705 metabolic reactions) and 3057 chemical species (including 1491 metabolites and 1566 proteins or protein complexes). It is 50% larger in scope than the first reconstruction. Part of the new data originated from the iIN800 (Nookaew et al.) reconstruction that includes lipid metabolism; in addition the team managed to find literature evidence for a large number of transporters that were originally identified by computational hypothesis.

As usual, the entire network is supplied in a single SBML file and is also available as a searchable database.

About this resource

This is a portal to the consensus yeast metabolic network as reconstructed from the genome sequence and literature. It is a highly annotated metabolic map that is periodically updated by a team of collaborators from various research groups. The first version of this reconstruction was published in Herrgård, Swainston et al. (2008) "A consensus yeast metabolic reconstruction obtained from a community approach to systems biology" Nature Biotechnol. 26, 1155-1160 (you can access that network here). A second version was released online; this third version is awaiting publication. We plan on continuing to update this resource towards a complete metabolic network of yeast. All versions will remain accessible for historical purposes, however it is highly recommended that you always use the latest one since that is the most up to date.

This effort started on the shoulders of a number of reconstructions of the metabolic network of yeast based on genomic and literature data that were published separately. (iMM904 and iLL672) However, due to the different approaches utilized in them, those earlier reconstructions had a significant number of differences. In addition they suffered from the use of non-standard names and overall they were not annotated with methods that are machine-readable. A community effort in 2007, led by the Manchester Centre for Integrative Systems Biology and the YSBN resulted in a consensus network representation of yeast metabolism, reconciling the earlier results. That effort is now ongoing under the leadership of the MCISB and with collaboration with colleagues under the UNICELLSYS FP7 project.


The network reconstruction is primarily assembled and provided as an SBML file enriched with MIRIAM-compliant annotations (which are embedded in the SBML through RDF). All small and macro- molecules are referenced to an authoritative database (e.g. Uniprot, ChEBI, etc.). All molecules and reactions are also annotated with appropriate publications that contain supporting evidence. Thus this network is entirely traceable and is presented in a computational framework. SBML is a format that is understood by a large number of software applications (see

While the SBML file is the most efficient computational resource for these data, casual users also need access to the network. That is provided by a searchable relational database accessed directly from this website. The database pages also allow readers to add comments to any chemical species or reaction. Such comments are taken into consideration by the team collating new versions of the network and can lead to corrections and additions to the network.

This reconstruction is provided in the following formats: