My work focuses on the development of formal methods in order to provide theoretical tools enabling the analysis of the dynamic in networks of interaction.
My researches led me to study self-organisation phenomena in such systems and to propose a framework in which describe, verify and simulate algorithms allowing the distributed assembling of structures specified by graphs. In collaboration with Cosimo Laneve from the University of Bologna, I developed in a second step the bioκ-calculus. This language allows the representation of a simplified and qualitative molecular biology based on interactions involving proteins and membranes. This part of my researches dealt in particular with exploring relevant definitions of equivalences between biological systems. These two aspects of my works rely mainly on the theory of process algebra and are naturally part of my PhD thesis (in french), written in the PPS lab (Université Paris 7) under the supervision of Vincent Danos.
I am currently working in the LIX and ISC laboratories in extending this effort in formalizing biological systems in a post-doctoral fellowship within the European project Morphex. The main objectve is to improve the understanding of the events involved in the morphogenesis of different kinds of organisms by focusing on the development of two sponges (Suberites domuncula and Lubomirskia baikalensis) and of the reproductive organs of a flowering plant (Arabidopsis thaliana). Again, this project deals with modelling issues in molecular biology. The challenge here is to succeed in unifying in a single approach both the description of the gene regulation networks involved in the morphogenesis and the events occuring at the cell level. The representation of these two levels of description calls for the design of dedicated tools and concerns more generally the study of complex systems. In this context, I am currently working with Leo Liberti (LIX) in adapting tools related to the operational research field and the mathematical programmation for soving inverse problems as well as the reformulation techniques which they are related to. The natural application of this work consists in enabling the reconstruction of biological networks from partial observed data.
In relation with this work on networks reconstruction, I also started a collaboration with Matthieu Latapy (LIP6 - Complex Networks) regarding the problematics raised from the study of dynamic graphs. In such a context of analysing large scale networks of interaction, the question of being able to extract rapidely a representative view of a given network in order to enable pertinant and valid anaysis of such a sample becomes more and more important. Beyond the quantitative aspect of the question, expressed as how fast we are able to retrieve a certain amount of existing links in the original graph, the qualitative perspective appears to be of paramount importance: how to characterize the bias induced by the techniques used to extract the samples. Those questions take place in a large thematic relying on the graph theory to propose new tools able to address those problems.