Épidémiologie évolutive des maladies infectieuses : approches théorique et statistique
Year: 2024
Type: dissertation
Abstract: Evolutionary epidemiology theory of infectious diseases aims to understand the interplay between epidemiological and evolutionary processes. This approach is particularly useful to study the transient dynamics of emerging pathogens, when epidemiological and evolutionary timescales overlap. For instance, it is essential for understanding the COVID-19 pandemic to account for the evolution of the virus and the succession of variants of concern. This adaptation can affect key phenotypic traits and erode our ability to mitigate epidemics. The advent of sequencing methods makes it now possible to collect genetic data and track the distribution of different strains across space and time. In this thesis, I combine evolutionary epidemiology theory and statistical inference using both epidemiological and genetic data to estimate pathogen phenotypes in public health and experimental microbiology. Throughout, my work relies on the analysis of deterministic models based on dynamical systems of ordinary differential equations.First, I focus on the rise of the SARS-CoV-2 Alpha variant in England and I explore which phenotypic traits drive its selective advantage. For this purpose, I develop a two-step approach based on SEIR epidemiological models. In the first step, before the emergence of the variant, I estimate the impact of the intensity of non-pharmaceutical interventions on the spread of the virus. In a second step, after the emergence of the variant, I exploit the slow-fast dynamics of eco-evolutionary processes to infer the phenotypic differences between the two competing lineages. I show that the selective advantage of the Alpha variant is likely to be driven by a higher transmission rate than by a longer infectious period. Second, I focus on an evolution experiment that tracked the epidemiological and evolutionary dynamics of the temperate bacteriophage lambda throughout its spread in a bacterial population of Escherichia coli. I develop a new inference approach to estimate the viral phenotypes at different stages of the epidemic -- including phenotypic traits very difficult to estimate otherwise. I model hidden processes such as lysis and lysogeny and fit this new model to an incomplete dataset. Third, I analyse how migration between host populations can affect the transient epidemiology and evolution of the pathogen. To do so, I track the transient evolutionary dynamics of the competition between two strains in an SIRS model. I show how migration can bias the quantification of the strength of selection and lead to misinterpretations about the real selective advantage of variants.These three projects allow me to develop new tools to exploit datasets that give access to an incomplete description (hidden processes and missing data) of the dynamics of a pathogen spreading and evolving in a heterogeneous host environment. Notably, I show that accounting for the pathogen structure among different compartments can be key to estimate model parameters, highlighting the importance of the availability of stratified data. This work emphasizes how the theoretical and statistical analysis of the joint epidemiological and evolutionary dynamics of infectious diseases can provide insights on the phenotypic evolution driving pathogen adaptation.
Source: theses.fr (ABES)
Author: Benhamou, Wakinyan
Language: English
Subfield: Genetics
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7090619125
Cites:
Cited by:
Related to: