Dynamical modelling of T cell inhibitory mechanisms in the immune response to cancer

Informations générales
Détails de la thèse/HDR
Anne Siegel
Rob de Boer
Inna Kuperstein
Anna Niarakis
Ioannis Xenarios
Directeur (pour les thèses)
Denis Thieffry
Morgane Thomas-Chollier
Résumé en anglais
Antibodies blocking the functions of T Cell co-inhibitory receptors
(checkpoint inhibitors) have become standard treatment for metastatic
melanoma. However, a mechanistic understanding of their harmful role
during anti-tumour responses has remained elusive.
The goal of my thesis is to better understand co-inhibitory processes in
CD4+ T cells and predict T Cell reaction to different immunotherapies
using computational approaches.
First, I have built detailed molecular maps of the processes occurring
during T cell activation based on information extracted from available
pathway databases and scientific literature. Next, this map was
translated into a more abstract regulatory graph, completed with logical
rules, to generate a complex discrete, dynamical model. The challenge
was then to properly model and analyse concurrent intracellular
processes, along with feedback and cross-talk mechanisms. To address
this challenge, I have developed and implemented a method to study the
behaviours of complex models through the semi-automatic extraction and
analysis of sub-models. I further automated the verification of such
(sub-)models based on biological specifications.
On a technical level, this work involved the conception of a Jupyter
(Python) notebook relying on a Docker image, in order to integrate the
whole panoply of tools required for these analyses, thereby ensuring
their reproducibility and reusability.
From a biological point of view, I have used the resulting model to
assess the impact of co-inhibitory receptors relatively to T Cell
activation. More specifically, using value percolation analysis, a
method derived from model reduction, I could show that our current model
corroborate the observed difference regarding the re-activation effects
of anti-CTLA4 versus anti-PD-1 immunotherapies in cancer