Description
Date depot: 19 juin 2023
Titre: Geometric deep learning for reconstructing conformation manifolds of biomolecules from cryo electron microscopy images
Directrice de thèse:
Slavica JONIC (IMPMC (EDITE))
Encadrant :
Jean FEYDY (Inria-Paris (ED-386))
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Intelligence artificielle
Resumé: Single-particle cryo electron microscopy (cryo-EM) allows 3D reconstruction of multiple conformations of purified biomolecular complexes from their 2D images. The elucidation of different conformations is the key to understanding the molecular mechanisms behind the biological functions of the complexes and the key to novel drug discovery. The standard cryo-EM data analysis procedures involve many rounds of 2D and 3D classifications to disentangle and interpret the combined conformational, orientational, and translational heterogeneity. Gradual conformational transitions give rise to many intermediate conformational states. Continuous conformational heterogeneity in cryo-EM data (a mixture of many intermediate conformational states), due to such gradual conformational transitions, is both an obstacle for high-resolution 3D reconstruction of different states and an opportunity to obtain the information about multiple coexisting states at once. 3D structures obtained by standard cryo-EM image analysis methods, involving classification and class averaging, hide conformations that are slightly different from the majority. Classical methods thus prevent the discovery of much less dominant conformational states, which could potentially be useful for new drug development. This PhD project will be focused on the development of cryo-EM image analysis methods for analyzing individual images without classification and averaging, based on geometric deep learning, which is better suited for data without an underlying Euclidean or grid-like structure such as protein structure. These new methods will be useful for obtaining biomolecular conformation manifolds from experimental cryo-EM data, with a potential future application to identifying most favorable conformations for virtual (in silico) drug screening.
Doctorant.e: Abid Eya