Skip to Main content Skip to Navigation

Modélisation mathématique des interactions entre pathogènes chez l’hôte humain : Application aux virus de la grippe et au pneumocoque

Abstract : Several pathogens have been suggested to interact with each other while circulating within human populations. These between-pathogen interactions may be synergistic, when one pathogen favours another, or antagonistic when one pathogen is detrimental to the other. Recent technological developments in the field of microbiology have created new opportunities for studying between-pathogen interactions within the human host, and particularly in the respiratory tract. These interactions may have dramatic consequences on the transmission dynamics of the implicated pathogens, and consequent public health impacts. However, despite some mechanistic hypotheses having been formulated, especially for the well-studied influenza-pneumococcus system, the underlying biological mechanisms are still poorly understood. This developing field raises numerous questions. From an epidemiological modelling perspective, how should these interaction mechanisms be formalized into models? How do these interactions impact the transmission dynamics and burden of the involved pathogens? Under which conditions, and with which methods can interactions be detected from ecological incidence data, classically reported from surveillance systems? The aim of this thesis is to address these questions using statistical and mathematical modelling tools, with a specific focus on the interaction between influenza and pneumococcus. While mathematical models have scarcely been used to address between-pathogen interactions, they are powerful tools that allow for a global approach by precisely formalizing the interactions at the individual scale and linking them to the population scale at which data are collected and phenomena observed. First, I conceived and developed a new agent-based model which simulates the co-circulation of two interacting pathogens in a human population. This model specifically formalizes between-pathogen interactions at the individual level, resulting in global dynamics at the population level. Notably, I demonstrated that different hypotheses regarding interaction mechanisms between influenza and pneumococcus lead to specific incidence dynamics and interaction burdens. Second, in order to construct in silico data mimicking surveillance data, I simulated a large number of interaction scenarios from the previous agent-based model. These simulated datasets were analysed using a variety of statistical and mathematical methods classically applied in between pathogen association studies. Results showed that all methods consistently detected between-pathogen associations as long as the simulated interaction strength remained above a threshold, which varied according to the method and the simulated interaction mechanism. Lastly, collaborating with the National Center for Pneumococcal Reference and Santé Publique France, we developed a new method to analyse between-pathogen interactions from incidence time series, based on the analysis of their seasonality patterns. By applying this method to French data of influenza-like illnesses and invasive pneumococcal diseases over the 2000-2014 period, we identified a small association, consistent with previous studies. The mathematical models developed and results presented in this thesis provide new understanding of the impact of between-pathogen interactions at the population level and the efficiency of available methods to assess them. Because the co-circulation of pathogens in populations is a complex system involving a large number of factors related to the pathogens, the host, and the environment, the development of mathematical models will be critical in the future. A better comprehension of these phenomena is of major importance as it may lead to new opportunities to reduce the public health burden of infectious diseases.
Document type :
Complete list of metadatas
Contributor : Abes Star :  Contact
Submitted on : Thursday, April 26, 2018 - 10:39:06 AM
Last modification on : Friday, March 6, 2020 - 3:30:12 PM
Long-term archiving on: : Tuesday, September 25, 2018 - 12:50:08 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01778961, version 1



Hélène Arduin. Modélisation mathématique des interactions entre pathogènes chez l’hôte humain : Application aux virus de la grippe et au pneumocoque. Santé publique et épidémiologie. Université Paris-Saclay, 2018. Français. ⟨NNT : 2018SACLV010⟩. ⟨tel-01778961⟩



Record views


Files downloads