Section: New Results

Summarized visualization of metabolic models

Participants : David James Sherman [correspondant] , Anna Zhukova.

In collaboration with Romain Bourqui and Antoine Lambert of the LaBRI, we defined new strategies for exploring whole genome metabolic models. There is an inherent tension between detail and understandability in these large networks: on the one hand, detailed description of individual reactions is needed for accurate simulation, but on the other hand, high-level views of reactions are needed for describing partways in human terms. We are defining knowledge-based simplification rules, that permit the user to factor similar reactions into one “generic” reaction in order to visualize a whole pathway or compartment, while maintaining the underlying model so that the user can later “drill down” to the specific reactions if need be. New layout rules implemented in the Tulip platform are used to draw the resulting networks in a familiar way.

In collaboration with Bruno Pinaud of the LaBRI, rule-based rewriting of metabolic models was used to define these simplifications using his PORGY software tool.