Description du sujet de thèse

Thesis subject : Analysis of the signal contained in the 3D structures of proteins using topological persistence
This thesis lies at the interface between evolutionary biology and algorithmic topology. It aims to optimise and extend topological data analysis methods, such as persistent homology, to study the information contained in the 3D structures of proteins. This thesis will tackle the problem of constructing the simplicial complex of a 3D protein structure using topological filtering that takes into account the physicochemical properties of amino acids and secondary structures. It will also apply multivariate and weighted persistent homology techniques to this context and refine the calculation of Wasserstein distance with probability measures on the barcodes of 3D structures, taking into account substitution mechanisms within sequences and protein sequence evolution models.

Contexte de travail

The successful PhD student will work in the Institut Camille Jordan (UMR5208), a mathematics laboratory headed by Véronique MAUME-DESCHAMPS (270 staff), and the LBBE laboratory (UMR5558), specialising in biometry and evolutionary biology, headed by Fabrice VAVRE (216 staff). The 2 laboratories are located in Villeurbanne on the La Doua campus of the Université Claude Bernard Lyon 1. This thesis is part of the ANR THERMADAPT project and will be supervised by Philippe MALBOS, the project's scientific manager, and Céline Brochier-Armanet from the LBBE.

Contraintes et risques

No specific risks identified.