Professor David Fleet (Google DeepMind, University of Toronto): 3d Structure and Motion from Single Particle Cryo-EM

Speaker: Professor David Fleet (Google DeepMind and University of Toronto)

Title: 3d Structure and Motion from Single Particle Cryo-EM

Video: Click to play

Abstract: One of the foremost challenges in structural biology concerns the inference of atomic-resolution 3D biomolecular structure from electron cryo-microscopy (cryo-EM).  The problem, in a nutshell, is a form of multi-view 3D reconstruction, estimating the 3D density of a biomolecule from large numbers of noisy images from a transmission electron microscope. The talk will outline the nature of the problem and discuss recent algorithmic advances. Notably, I’ll focus on the key problem of heterogeneity, in which the imaged particles are non-rigid due to conformational dynamics or compositional variation.

Bio: David Fleet is a Principal Scientist at Google DeepMind (since 2020) and a Professor of Computer Science at the University of Toronto (since 2004).  His research interests, at one time or another, have spanned computer vision, image processing, machine learning, visual neuroscience, and computational biology.  He has won research awards for papers at CVPR, ICCV, NeurIPS, BMVC, and UIST. He received the Alfred P Sloan Research Fellowship in 1996, and the Koenderink Prize in 2010 for work with Michael Black and Hedvig Sidenbladh.  With Ali Punjani his work on cryo-EM received the 2021 Paper of the Year award from the Journal of Structural Biology.  He served as Associate Editor of IEEE Transactions on  PAMI (2000-2004), as Program Co-Chair for CVPR (2003) and ECCV (2014), and as Associate Editor-In-Chief for IEEE Transactions on PAMI from 2005 to 2008. He was Senior Fellow of the Canadian Institute of Advanced Research from 2005 to 2019, and currently holds a Canadian CIFAR AI Chair at the Vector Institute for AI.