Systems biology: analysing data and modelling interactions

We address the question of constructing accurate models of biological systems with respect to available data and knowledge. The availability of high-throughput methods in molecular biology has led to a tremendous increase of measurable data along with resulting knowledge repositories, gathered on the web (e.g. KEGG,MetaCyc, RegulonDB). However, both measurements as well as biological networks are prone to incompleteness, heterogeneity, and mutual inconsistency, making it highly nontrivial to draw biologically meaningful conclusions in an automated way. Based on this statement, we develop methods for the analysis of large-scale biological networks which formalize various reasoning modes in order to highlight incomplete regions in a regulatory model and to point at network products that need to be activated or inactivated to globally explain the experimental data. We also consider small-scale biological systems for a fine understanding of conclusions that can be drawn on active pathways from available data, working on deducible properties rather than simulation. Corresponding disciplinary fields are model checking, constraint-based analysis and dynamical systems.

Participants

• Permanent members. J. Bourdon (Assistant professor, Nantes, on leave at INRIA), M. Le Borgne (Assistant professor, Rennes), F. Morrews (ingeneer, INRA), J. Nicolas (Research director, Inria), A. Siegel (Research scientist, CNRS).

• Associate members (researchers from other labs that also have an office in the team) : O. Radulescu (Professor, Montpellier),  N. Théret (research director, Inserm).

• Post-docs. F. Corblin, N. Le Meur, S. Blachon (2007-09).

• Ph-D students. P. Blavy (2006-10), A. Floeter (2003-06), J. Gruel (2006-09), C. Guziolowski (2006-09), P. Veber (2004-07).

Short abstract

We first focus on large scale systems. We have developed a formalism to reason automatically over large-scale biological networks, in order to detect inconsistencies between the knowledge and data, giving hints on lacks in knowledge and new potential experiments. After a manual step of model correction, this allows deducing information on non observed products.

To implement this strategy, we mainly describe systems with influence graphs and confront them with the information provided by DNA chip data. We have proposed a formal coding of an intuitive rule telling that the variations of any product must be explained by the variation of at least one regulator: we performed theoretical studies to overcome the range of viability of this rule (Siegel et al, biosystems 2006; Radulescu et al, Roy. Soc. Inter. 2006).

This coding has led us to solve systems of constraints on signs of variations, a NP hard problem, that we tackled with both decision diagrams techniques (Veber et al, complexUs 2005) and dedicated constraint solvers such as answer set programming (ASP), a recent paradigm that gathers the expressivity of non monotonic logics with the efficiency of SAT solvers and constrain systems solvers (Veber et al, BMC bioinformatics 2008).

Altogether, this allowed to integrate automatic reasoning functions in order to highlight incomplete regions in a regulatory model and reasons over which products in the network need to be activated or inactivated in order to globally explain the experimental data. Finally, a software tool (Cytoscape plugin) was designed to friendly perform such a global qualitative analysis (Guzioloswki et al, BMC genomics, 2009).

We validated our methods with the correction of E.Coli network extracted from the bank RegulonDB (Guziolowski et al, JBBC 2006 and CIBB 2008), by tackling the question of network inference for E.Coli and yeast (Veber et al, BMC bioinformatics, 2008), and by modeling of the cell-cycle perturbations induced by a chimeric oncogene (EWS-FLI1) responsible for pediatric bone cancer (Ewing sarcoma) (Blachon et al, BMINNT, 2009).

In parallel, we tackled the question of accurate dynamical modeling for small-scale biological systems. The general question we addressed was to understand which conclusion on active pathways can be deduced from available time-series data. A central underlying hypothesis is that dynamics of biological networks is hierarchical, involving many separated time scales. Therefore, we thus have developed a mathematical methodology allowing us to take advantage of the hierarchical nature of the system. This methodology relies on model reduction and comparison techniques, within and between various levels of descriptions of biological networks: already mentioned large-scale qualitative networks (Siegel et al, biosystems 2006; Radulescu et al, Roy. Soc. Inter. 2006), stochastic processes (Raduescu et al, TSI 2007), partial and ordinary differential equations (Gorban et al, IET syst. biol. 2007, .

With this general approach were considered several problems in physiology and pathology. We studied models of cellular signaling such as NF-kappaB signaling network (Radulescu et al, BMC syst. bio. 2008) and the effect of the protein ADAM12 in signaling and hepatic fibrosis process (Gruel et al, BMC research Notes, 2009). We considered decision making and development systems for drosophila (Manu et al, PLOS biology and PLOS comput. biol., 2009). We also considered the modeling of fatty acids metabolisms and its regulations, and we used a hierarchy of models to deduce that some pathways were active in mutant mice, suggesting the existence of an unknown regulator for these pathways (Blavy et al, JTB, 2009).

Review papers

• Christophe Lavelle, Hugues Berry, Guillaume Beslon, Francisco Ginelli, Jean-Louis Giavitto, Zoi Kapoula, André Le Bivic, Nadine Peyrieras, Ovidiu Radulescu, Adrien Six, V´eronique Thomas Vaslin, and Paul Bourgine. From Molecules to Organisms: Towards Multiscale Integrated Models of Biological Systems. Theoretical Biology Insights, 1:13–22, 2008.

• A. Siegel, C. Guziolowski, P. Veber, O. Radulescu, M. Le Borgne Optimiser un plan d'expérience à partir de modèles qualitatifs?,  BioFutur (275), 2007, 27-31

Methodological publications

Constraints-based approaches for the confrontation of variation data with influence graphs

• Correction under inconsistency. M. Gebser, C. Guziolowski, M. Ivanchev, T. Schaub, A. Siegel, P. Veber and S. Thiele Repair and Prediction (under Inconsistency) in Large Biological Networks with Answer Set Programming KR'2010 - International Conference on the Principles of Knowledge Representation and Reasoning, Toronto, 2010

• Software development.  C Guziolowski, A Bourdé, F Moreews, and A Siegel. BioQuali Cytoscape plugin: analysing the global consistency of regulatory networks. BMC Genomics, 26(10):244, 2009.

  Sofware : Bioquali Cytoscape plugin , Web access

• Introducing complex-formation constraint rules. C. C. Guziolowski, J. Gruel, O. Radulescu, , and A. Siegel. Curating a large-scale regulatory network by evaluating its consistency with expression datasets, Selected peer-reviewed version of CIBB 2008 : computational Intelligence Methods for Bioinformatics and Biostatistics. volume 5488 of Lecture Notes in Computer Sciences, pages 144–155. Springer-Verlag, 2009.

• Application to E. Coli and S. cerevisiae network inference. Philippe Veber, Carito Guziolowski, Michel Le Borgne, Ovidiu Radulescu, and Anne Siegel. Inferring the role of transcription factors in regulatory networks. BMC Bioinformatics, 9, 2008.

• Summary of mathematical features. Anne Siegel, Carito Guziolowski, Philippe Veber, Ovidiu Radulescu, and Michel Le Borgne. Qualitative response of interaction networks: application to the validation of biological models. In 6th International Congress on Industrial and Applied Mathematics. ICIAM 07, Zurich Suisse, 2007.

• Confrontation of large-scale network and data. Proof of concepts. Carito Guziolowski, Philippe Veber, Michel Le Borgne, Ovidiu Radulescu, and Anne Siegel. Checking Consistency Between Expression Data and Large Scale Regulatory Networks: A Case Study. The Journal of Biological Physics and Chemistry, 7:37–43, 2007

• Influence graph reduction and constraint solving with ternary decision diagrams.  P Veber, M Le Borgne, A Siegel, S Lagarrigue, and O Radulescu. Complex qualitative models in biology: A new approach. Selected peer-reviewed version of ECCS’05. Complexus, 2(3-4):140 – 151, 2006.

• Modelling steady states transitions. O. Radulescu, S. Laguarrigue, A. Siegel, M. Le Borgne, and P. Veber. Topology and linear response of interaction networks in molecular biology. Journal of The Royal Society Interface, 3(6):185 – 196, 2006.

• Constraints raised by steady states. A. Siegel, O. Radulescu, M. Le Borgne, P. Veber, J. Ouy, and S. Laguarrigue. Qualitative analysis of the relation between dna microarray data and behavioral models of regulation network. BioSystems, 84:153–174, 2006.

Data analysis

• Si-RNA. A. Bankhead, I. Sach, C. Ni, N. Le Meur, M. Kruger, M. ferrer, R. gentleman, C. Rohl  Knowledge based identification of essential signaling from genome-scale siRNA experiments., in "BMC Syst Biol", vol. 3, 2009, 80. 

• Lethality data. Nolwenn Le Meur and Robert Gentleman. Modeling synthetic lethality. Genome Biology, 9:R135, 2008.

• Metabolic data. Threshold extraction in metabolite concentration data A Floeter, J Nicolas, T Schaub, and J Selbig Bioinformatics 20:1491-1494. 2004.

Robustness, reduction and time-scales hierachies in biological networks

• Reduction of stochatic models. A.Crudu, A.Debussche, and O.Radulescu. Hybrid stochastic simplifications for multiscale gene networks . BMC Systems Biology, 2009.

• Reduction of continuous reaction models. Alexander Gorban and Ovidiu Radulescu. Dynamic and static limitation in multiscale reaction networks, revisited. Chemical Engineering Science, 34:103–173, 2008.

• NFkappaB application. Ovidiu Radulescu, Alexander Gorban, Andrei Zinovyev, and Alain Lilienbaum. Robust simplifications of multiscale biochemical networks. BMC Systems Biology, 2:86, 2008.

• Modèles stochastiques. Ovidiu Radulescu, Aurélie Muller, and Alina Crudu. Théorèmes limites pour des processus de Markov à sauts. Synthèse des résultats et applications en biologie moleculaire. TSI (Technique et Science Informatiques), 26:443–469, 2007.

• Robustness. Alexander Gorban and Ovidiu Radulescu. Dynamical robustness of biological networks with hierarchical distribution of time scales. IET Systems Biology, 2007.

• Hierarchies. O Radulescu, A Gorban, and A Zinovyev. Hierarchies and modules in complex biological systems. In European Conference on Complex Systems - ECCS’06, 2006

Model construction and applications

Regulation of metabolic networks

• Regulations of fatty acid metabolic pathways. P. Blavy, F. Gondret, H. Guillou, S. Lagarrigue, P.G.P. Martin, J. van Milgen, O. Radulescu, and A. Siegel. A minimal model for hepatic fatty acid balance during fasting: Application to ppar alpha-deficient mice. Journal of Theoretical Biology, 2009.

• Mammary metabolism in lactating dairy cows. Sophie Lemosquet, Jérémie Bourdon, Jocelyne Guinard-Flament, Anne Siegel, and Jaap Van Milgen. A generic stoichiometric model to analyse the metabolic flexibility of the mammary gland in lactating dairy cows. In 7th International Workshop: Modelling Nutrition Digestion and Utilization in Farm Animals, 2009.

• Fatty acid metabolism : network abstraction. Pierre Blavy, Florence Gondret, Herv´e Guillou, Sandrine Lagarrigue, Pascal Martin, Ovidiu Radulescu, Anne Siegel, and Jaap Van Milgen. A minimal and dynamical model for fatty acid metabolism in mouse liver. In JOBIM : Journ´ees Ouvertes en Biologie Informatique et Mathématiques, 2008.

Signalling networks

• Sea Urchin. R. Bellé, S. Prigent, A. Siegel and P. Cormier,  Model of cap-dependent translation initiation in sea urchin. A step towards the eukaryotic translation regulation network Molecular Reproduction and Development 77(3), 2010, 257-264

• Effets EGF and TGF-b with multiclocks models. N. Le Meur, J. Gruel, M. Le Borgne, N. Theret. Modeling the influence of EGF and TGF-b pathways in tumor progression of hepatocellular carcinoma., in "Asian Pacific Association for Study of the Liver, Hong- Kong", vol. 3, 2009, 64, FP107,

• Effect of ADAM 12 in TGFbeta signalling pathways. Jérémy Gruel, Michel LeBorgne, Nolwenn LeMeur, and Nathalie Th´eret. In silico investigation of adam12 effect on tgfb receptors trafficking. BMC Research Notes, 2009.

• Analysis of CGH data in Ewing tumor regulation network S. Blachon, G. Stoll, C. Guziolowski, A. Zimovyev, E. Barillot, A. Siegel, and O. Radulescu. Method for relating inter-patient gene copy numbers variations with on Biomedical Informatics and Intelligent Approaches in the Support of Genomic Medicine (BMIINT), volume 475 of CEUR Workshop proceedings, pages 72–87, 2009.

• NFkappaB. Ovidiu Radulescu, Andrei Zynoviev, and Alain Lilienbaum. Model reduction and model comparison for nfkb signaling. In FOSBE’07: Foundations of Systems Biology in Engineering, Stuttgart, 2007.

Drosophila development

• Biological paper S.S. Manu, A.V. Spirov, V.V. Gursky, H. Janssens, A.R. Kim, O. Radulescu, C.E. Vanario-Alonso, D.H. Sharp, M. Samsonova, and J. Reinitz. Canalization of gene expression in the Drosophila blastoderm by gap gene cross regulation. PLoS Biology, 7(3), 2009.

• Computational paper. S.S. Manu, A.V. Spirov, V.V. Gursky, H. Janssens, A.R. Kim, O. Radulescu, C.E.Vanario-Alonso, D.H. Sharp, M. Samsonova, and J. Reinitz. Canalization of gene expression and domain shifts in the Drosophila blastoderm by dynamical attractors. PLoS Computational Biology, 5(3), 2009.

PhD and habilitation thesis

• Pierre Blavy. Identification des éléments clef du métabolisme des lipides et de leurs régulateurs. PhD thesis, Agrocampus, march 2010.

• Jérémy Gruel. From biological data to molecular modelisation; application to TGF-beta signaling and hepatic fibrosis, Ph. D. Thesis, Université de Rennes1, 2009.

• Carito Guziolowski.  Analysis of Large-Scale Biological Networks with Constraint-Based approaches over Static Models.  PhD Thesis, University of Rennes 1, France, Jan 2010  (version electronique)

• Anne Siegel. Analyse de systèmes dynamiques par discrétisation. Exemples d’applications en théorie des nombres et en biologie moléculaire. Habilitation thesis, Université Rennes 1, 2008. (version electronique)

• Ovidiu Radulescu. Habilitation à diriger des recherches: Modèles mathématiques de la complexité en biologie moléculaire et en m´ecanique des fluides. Habilitation thesis, Université de Rennes 1, 2006. (version electronique)

• Philippe Veber. Modélisation grande échelle de réseaux biologiques : vérification par contraintes booléennes de la cohérence des données. PhD thesis, Université Rennes 1, 2007 (version electronique )

• André Floeter. Analysing biological expression data based on decision tree induction. PhD thesis, Universit´e de Rennes 1 / Postdam University, january 2006.