Date : 14/02/2011

Laboratory
Ecology & Evolution
UMR 7625 CNRS- UPMC-ENS
Ecole Normale Supérieure
46 rue d'Ulm 75230 PARIS
Director : Minus Van Baalen
Main discipline : Epidemiology
Lab's website


PhD Supervisor
Bernard Cazelles
email : Cet e-mail est protégé contre les robots collecteurs de mails, votre navigateur doit accepter le Javascript pour le voir
phone : +33 1 44 32 38 76

Subjects
1.: Epidemiological and Evolutionary dynamics
2.: Influenza Epidemics

Tools-Methodologies:
1.: Stochastic models
2.: Individual Based Models
3.: Bayesian inference

Summary of lab's interests

The PhD project will be done in collaboration between two lab "Ecologie & Evolution" and "Epidémiologie, Systèmes d'Information, Modélisation", and between two teams of these labs with Bernard Cazelles in the team "Eco-Evoution Mathematique" and Pierre-Yves Boëlle in the team "Epidémiologie des Maladies Infectieuses et Modélisation". The first laboratory's research interests are in evolutionary ecology, with a strong focus on feedbacks between ecological and evolutionary processes. Main research axes are eco-evolutionary modeling, epidemiology, evolutionary microbiology and immunology, evolution of sociality, integrative ecology. The second laboratory has a high international in quantitative epidemiology. The specificity of the research group is the preponderant use of mathematics models, the statistics, the computer sciences and its use of information system technologies to lead its work in the domain of epidemiology.

Summary of project

The main objective of the thesis is to contribute to the understanding of the complex mechanisms involved in the invasion and persistence of a phenotypically distinct strain of the influenza virus by developing a mathematical framework that accounts for the probabilistic aspects related to these processes. This stochastic framework would account for changes at the phenotypic level, both gradual and epochal. The developed framework will be applied to flu epidemics in temperate areas and its development will be based on data from the "Réseau Sentinelles" (www.sentiweb.org, INSERM S707) for the study of recurrent persistence and co-circulation and of the COPANFLU cohort (coordinated by F. Carrat) for the study of emergence related mechanisms. The serologic and genetic data from the CNR should also be employed. The points that we wish to assess are related to the following questions: Which are the persistence conditions for a new emerging strain? Which are the interactio n mechanisms between co-circulating strains and in which circumstances does the new variant replace the dominant circulating strain? Could the co-circulation of several distinct strains modify the pathogen dynamic and render difficult the prediction of epidemics? The main aspect of the PhD work wiil consist in building consistent mathematical formulations of multi-strain interactions, which have to be biologically relevant and also computationally tractable in order to be applied to real complex situations. We propose to start with a stochastic framework with historical based formulation that can incorporate more than two different strains of a pathogen and their interactions and an age structure for the host population. It will also be applied to explore the conditions supporting the non-emergence of the pandemic nao A/H1N1 influenza and its non-ability to replace the dominant circulating strain A/H3N2. An important point will be related to the estimation of key epidemiological parameters using recent inference techniques based on particle filters. These techniques using the state-space framework with some specific equations specifically describing the observation process allow the estimation of models parameters (their means but als o their distributions) from uncertain and partially observed data. Moreover these techniques permit directly the computation of model likelihoods allowing a comparison of different models and hence testing the relevance of underlying biological assumptions.

Interdisciplinarity of the project

As new pathogens continue to emerge in humans, wildlife and domestic animals and plants. Understanding the retroactive loops between epidemiological and evolutionary mechanisms of rapidly evolving pathogens, often characterized by several phenotypically distinct strains, is one of the most challenging problems in the research field of infectious diseases. In this context, mathematical modeling provides a theoretical framework for improving our understanding of these evo-epidemiological complex systems, which are perpetually evolving. Mathematical models can be used to evaluate public health interventions for preventing and controlling the spread of infectious diseases and also to propose innovative methods for detecting and the characterizing emerging and re-emerging pathogens. Nevertheless, despite a well-posed conceptual framework for evaluating the spread of infections available in theoretical epidemiology, there are few models that have been developed for multi-strain s diseases where different phenotypic strains of the pathogens coexist. Essential questions still remain on the specific biological mechanisms involved as well as on the appropriate modeling approaches for describing such complex eco-epidemiological models and estimating their key parameters. An emblematic example of these evo-epidemiological complex systems is the influenza epidemics. At each pandemic of human influenza A, the new stain replaced the dominant circulating strain. This was observed in 1918 when the sub-type A/H1N1 replaced the A/H3N8 sub-type that has been present since 1900. A/H1N1 was then been replaced in 1957 by a new emerging sub-type, the A/H2N2, which gave way at his turn in 1968 to the Hong-Kong strain belonging to the sub-type A/H3N2. Since 1977, the coexistence of strains of two sub-types of influenza A has been observed, after an accidental reintroduction of a A/H1N1 variant. Nevertheless in 2009 pandemic due to the new animal origin (nao) A/H1N1 virus has not success in the remplacement of sub-type A/H3N2. Then the 2009 pandemic and the data collected since 2009 appear as a unique opportunity for comparing theoretical predictions to observations for a better understanding of coexistence and replacement mechanisms, which form the core of this PhD project.