Section: New Results

Models of carbon metabolism in bacteria

Kinetic models capture the dynamics of the large and complex networks of biochemical reactions that endow bacteria with the capacity to adapt their functioning to changes in the environment. In collaboration with Matteo Brilli and Daniel Kahn (Inra and Université Claude Bernard in Lyon), we previously developed an approximate model of central metabolism of E. coli, using linlog kinetics, and estimated the parameter values from metabolomics, transcriptome, proteomics data sets, as described in an article published in Bioinformatics in 2011. The results of this study revealed the fundamental role played by the identifiability of the model parameters, an issue often overlooked in systems biology. This prompted us for a thorough investigation of the concepts of structural identifiability (in presence of perfect, idealized data), practical identifiability (in presence of noisy and limited amounts of data), and the relations between the two. In addition, we looked into the implications of this analysis for the reduction of nonidentifable to identifiable models. While having a solid mathematical basis, the study was tailored to the actual experimental practice, and resulted in a practical model reduction method that improves upon our previous approach in case of large measurement noise. This study, and the results from its application to both in-silico case studies and state-of-the-art datasets, were reported in a paper that appeared in the Journal of Mathematical Biology this year [4] . Although the theoretical development has focused on linlog models and related classes of approximate kinetic models, it is important to note that the results also bear on more general classes of nonlinear models of metabolism.

A second line of work is based on the use of classical kinetic models that are, in comparison with the above-mentioned linlog models, much reduced in scope (the focus is on the metabolic and genetic regulation of the glycolysis pathway) and granularity (individual reactions are lumped together). The models, developed by Delphine Ropers, are being calibrated using experimental data from IBIS group and the group of Jean-Charles Portais at Inra/INSA in Toulouse, and will be used to understand some key mechanisms in the adaptation of E. coli to the exhaustion of glucose. The PhD thesis of Manon Morin, in the framework of a collaboration supported by a Contrat Jeune Scientifique Inra-Inria, will further develop these research directions. In the framework of their PhD theses, Stéphane Pinhal and Valentin Zulkower also study specific aspects of carbon metabolism, using both models and experimental data. In parallel, we collaborate with Myriam Ferro at CEA in Grenoble to investigate how state-of-the-art measurements of the absolute concentrations of enzymes in E. coli can be integrated with other high-throughput data sets and kinetic models. The first results of this collaboration were accepted for publication in Molecular and Cellular Proteomics early 2014.