Description
Date depot: 28 octobre 2022
Titre: Modelling hidden causes in disease progression
Directeur de thèse:
Hervé ISAMBERT (PC_Curie)
Encadrante :
Barbara BRAVI (Dept. of Mathematics)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Sciences de l’information et sciences du vivant
Resumé: A complete understanding of the cellular and molecular determinants of complex disease progression is crucial to the identification of effective targets for treatment and prognostic markers of patients’ response. However we are far from achieving such understanding, particularly for non-communicable diseases, like most cancer and neurodegenerative diseases, which are leading causes of death on a global scale, or auto-immune diseases, like Type 1 diabetes and multiple sclerosis, which are long-term chronic illnesses. Yet, the last decade has brought along a series of powerful novel technologies that are able, on the one hand, to produce unprecedented large amounts of live-cell imaging and gene expression data at single cell resolution and, on the other hand, to extract information from these data through statistical and machine learning, giving hope to significantly improve our comprehension of disease progression. The objective of this PhD project is to leverage the potential of such information-rich technologies to build predictive models that are amenable to a causal and mechanistic interpretation, and that can thus help uncover functional processes in healthy cellular systems and their gradual disruption under disease progression.
The Isambert lab recently developed novel causal inference methods and tools (https://miic.curie.fr) to learn cause-effect relationships in a variety of biological or clinical datasets, from single-cell transcriptomic and genomic alteration data (Verny et al 2017, Sella et al 2018, Desterke et al 2020) to medical records of patients (Cabeli et al 2020, Sella et al 2022, Ribeiro Dantas et al 2022) These machine learning methods combine multivariate information analysis with interpretable graphical models (Li et al 2019, Cabeli et al 2021, Ribeiro Dantas et al 2022) and outperform other methods on a broad range of benchmarks, achieving better results with only ten to hundred times fewer samples.
The objectives of the present PhD project, in collaboration with Barbara Bravi’s team at Imperial College London, are, first, to parametrize reconstructed causal networks in order 1- to predict the course of a disease from early temporal information and, second, to generate synthetic data, which will then be used to improve causal inference methods through an iterative ‘adversarial’ model training approach. These novel advanced causal inference methods will then be applied to analyze high-through put cell biology data, such as 1- features extracted from time-lapse images of i) tumour-on-chip cellular ecosystems in collaboration with Maria Carla Parrini (Inst Curie) and ii) differentiating hematopoietic stem cells in collaboration with Leïla Perié (Inst Curie) and 2- single-cell transcriptomic data of i) breast cancer under treatment in vitro in collaboration with Luca Magnani (Imperial College) and ii) differentiating hematopoietic stem cells from the Perié lab (Inst Curie).
Doctorant.e: Lagrange Nikita