Home  > News from Ibis > Training and job opportunities

Offer for PhD position (closed)

Modeling and identification of large models of regulatory networks in bacteria


Much of the functioning of an organism consists in responding to changes in the environment. This adaptation is accomplished by regulatory networks, consisting of genes, mRNAs, their products, metabolites, and all the mutual regulatory interactions. Understanding the functioning of complex regulatory networks is therefore a primary goal of fundamental research in biology, but also the basis for numerous applications in biotechnology and human health. This is achieved through the combination of experimental data with mathematical modeling and computer-aided analysis techniques, in the context of a broader movement called systems biology.

Modern experimental techniques have led to an increase of the availability and quality of high-throughput data characterizing the state of an organism over time. This has stimulated the development of larger kinetic models that provide a broader and more refined description of the functioning of an organism. The quantitative predictions obtained with the models rely on the determination of non-measurable parameters from experimental data. This parameter estimation problem is hampered by the specificities of biological systems : incomplete knowledge of the network structure and molecular mechanisms ; noisy, indirect, heterogeneous, and partial observations ; stochastic phenomena when few molecules are involved, dynamics on different time-scales. As a consequence, model identification is underdetermined and remains a major bottleneck in quantitative systems biology.

Currently, available methods work reasonably well for small-to-moderately-sized networks, but the limited scalability may impair their application to larger networks integrating more diverse and complex biological processes. This motivates the subject of this PhD thesis : the development, evaluation, and application of identification methods optimally suited for the kinetic modeling of large-scale biological regulatory networks. The research activity will be carried out within the IBIS group at the INRIA Grenoble - Rhône-Alpes center located in Montbonnot.

We will particularly focus on approaches that split the model identification into subproblems, corresponding to different network modules, so as to reduce the problem complexity. These methods will be developed in the context of a real biological application : the study of the regulatory networks involved in the adaptation of the enterobacterium Escherichia coli to changes in the environmental carbon sources. These networks are investigated by Johannes Geiselmann (member of IBIS) at the Microorganism Pathogeny and Adaptation Laboratory of the University Joseph Fourier (LAPM, Grenoble) and Jean-Charles Portais at the Bioprocess and Biosystems Engineering Laboratory of the University Paul Sabatier (LISBP, Toulouse).

The above core problem in the identification of biological regulatory networks will be addressed in the context of other aspects of a more general modeling problem, including :
-  Formulation of kinetic model and model reduction. The mathematical tools that are most appropriate for the description of the regulatory mechanisms are chosen on the basis of the later usage of the model and on the quality of the available experimental data. In parallel, model reduction addresses the choice of the right compromise between accuracy and complexity of the model.
-  Data analysis and signal processing. The problem is to handle datasets of ever-increasing size and process raw experimental measurements into informative datasets that can be related to variables in the chosen modeling framework.
-  Model discrimination and experiment design. Given that the identification process does not usually result in a single solution, that is, a single quantitative model, the problem is to propose novel experiments that are expected to yield a maximum of information for refining the estimates of parameter values or discriminating between alternative model structures.

Depending on the profile and interests of the candidates, the project may include one or more of the above aspects, and could also involve a contribution to the realization of the experiments at LAPM.

Competences and requirements

We are looking for a computer scientist, mathematician or physicist with experience in modeling and identification, preferably of biological systems, or a biologist with a solid foundation in mathematics. Candidates interested in conducting experiments are welcome.

Relevant references

-  Kitano H., eds, Foundations of Systems Biology, MIT Press, Cambridge, MA, 2001.
-  Szallasi Z. et al., eds, System Modeling in Cellular Biology : From Concepts to Nuts and Bolts. MIT Press, Cambridge, MA, 2006.
-  M. Ashyraliyev, Y. Fomekong Nanfack, J.A. Kaandorp, J.G. Blom (2009), Systems biology : Parameter estimation for biochemical models, FEBS J., 276(4):886-902.
-  I. Cantone, L. Marucci, F. Iorio, M. Bansal, V. Belcastro, A. Ricci, S. Santini, M. di Bernardo, D. di Bernardo, M.P. Cosma (2009), A yeast synthetic network for in vivo assessment of reverse-engineering and modeling approaches, Cell, 137:1-10.
-  E.J. Crampin (2006), System identification challenges from systems biology, Proc. 14th IFAC Symposium on System Identification (SYSID 2006), Newcastle, Australia, 87-96.
-  M. Rodriguez-Fernandez, P. Mendes, J. R. Banga (2006), A hybrid approach for efficient and robust parameter estimation in biochemical pathways, BioSystems, 83(2-3):248-265
-  O. Kotte, M. Heinemann (2009), A divide-and-conquer approach to analyze underdetermined biochemical models, Bioinformatics, 25:519-525.


This position is offered at the INRIA Grenoble - Rhône-Alpes Research Unit of INRIA, located near Grenoble and Lyon. The Unit includes more than 600 people, within 34 research teams and 10 support services.

Application and additional information

Application deadlines, details of the application procedure and additional information on Ph.D. positions at INRIA (eligibility, duration, salary...) are reported on the INRIA official website.

Interested candidates should contact :

Delphine Ropers

Hidde de Jong