Metexplore: Handling Genome Scale Metabolic Networks Online

Since 2009 the MetExplore web server (www.metexplore.fr) has offered access to more than 200 metabolic networks representing a large variety of organisms and coming from various databases (e.g., BioCyc, KEGG, biomodels…). A key feature of MetExplore is its provision of a single framework that allows analysis of omics datasets within the context of these networks. We created a fully online interface (from data import to network visualization) thus eliminating any need for software installation and file exchanges. Main functions include graph modelling, SBML file import and export and Cytoscape exports. Constraint based modelling (FBA, FVA …) on in silico engineered models are possible through computation in a newly developed Java framework (FlexFlux).

Besides making available public metabolic network reconstructions, MetExplore allows to import additional networks as SBML files as well as to build new metabolic networks via a user interface. These draft networks are stored in a private section of the database. MetExplore provides an online collaborative framework to annotate metabolic network reconstructions. Owners of a reconstruction can share it with other registered users to split the curation effort. Newly created and public available networks can be exported as SBML files. 

MetExplore is capable of extracting a variety of sub-networks from within an initial larger network, based on individual pathways, or modules responding to environmental challenges. MetExplore tackles sub-network identification in two ways.  A first approach filters for selected features, e.g. certain metabolites or reactions either known to be involved in particular responses, or inferred to be involved through association with others that are identified.  Secondly, application of graph algorithms to the network structure can identify metabolic paths between identified metabolites.

MetExplore originality also relies on its web based visualization of metabolic networks. It is possible to visualize and explore an entire network or to display specific parts of a network from a selection of reactions and/or pathways. Moreover, omics data can be visualized using color mapping on nodes. In order to go further in visual inspection, these networks can also be exported to Cytoscape.

The MetExplore architecture allows a flexible analysis of integrated omics data within the context of metabolic networks. Transcriptomic and metabolomic data can be mapped onto networks available in MetExplore. It has been successfully used to interpret metabolomics data obtained from a range of research areas including microbiology, pharmacology, food toxicology and parasitology