Joseph Nardin—Gennequin
I am a PhD student in mathematics and bioinformatics at the Institut Camille Jordan and the Laboratoire de Biologie et Biométrie Évolutive, which are both part of Université Claude Bernard Lyon 1.
Research
My research lies at the interface of computational algebraic topology, spectral geometry, deep learning, and structural bioinformatics. I am interested in developing faithful representations of protein structures with topological and spectral invariants, and leveraging these representations through machine learning approaches. In particular, I focus on applications of these methods to predictive biological modelling tasks such as phylogeny reconstruction, life history traits prediction (optimal growth temperature and pressure) and structural classification. Other applications that I would like to work on in the future include function prediction, binding site location prediction, and the analysis of protein-protein interactions.
My work is part of the Deltafold research project, led by my two supervisors: Prof. Philippe Malbos and Prof. Céline Brochier—Armanet, and is funded by the project ANR Thermadapt (22-CE02-0027).
Publications
- Supervised classification of protein structures using topological persistence with Deltafold. With G. Ciuperca, C. Brochier—Armanet and P. Malbos. Submitted to Proteins — Structures, function and bioinformatics.
- Disclose protein structures through topological persistence: a review. With L. Bou Dagher, F. El Chami, C. Brochier—Armanet and P. Malbos. Preprint.
- Spectral geometry captures phylogenetic fingerprints in protein structures. With C. Brochier—Armanet and P. Malbos. In preparation.
- Topological Deep Learning for protein structure classification. With N. Ochonisky—Girand, C. Brochier—Armanet and P. Malbos. In preparation.
Conferences and seminars
I have presented talks and posters at the following conferences:- Talk at the Geometry, Topology and Machine Learning conference, Paris, May 2026: Topological deep learning for protein structure classification.
- Talk at the Geometry and Machine Learning conference, Paris, May 2025: Geometry of protein structures for phylogeny and classification.
- Talk at the Journées de rencontre des doctorants du réseau RTop, Amiens, April 2025: Persistance topologique et géométrie spectrale pour l'analyse des structures de protéines.
- Poster at the Nancy Computational Structural Biology Meeting, Nancy, July 2024: Quantification of structural properties of proteins with topological data analysis.
Teaching
I have previously taught at the Institut de Sciences Financières et d'Assurances, the UFR Biosciences, and the math department of the Faculté des Sciences of Université Claude Bernard Lyon 1. Details and course materials are available here in French only.