Timezone: »

 
Poster
Amortized Inference for Heterogeneous Reconstruction in Cryo-EM
Axel Levy · Gordon Wetzstein · Julien N.P Martel · Frederic Poitevin · Ellen Zhong

Tue Nov 29 02:00 PM -- 04:00 PM (PST) @ Hall J #225
in Poster Session 2 »

Cryo-electron microscopy (cryo-EM) is an imaging modality that provides unique insights into the dynamics of proteins and other building blocks of life. The algorithmic challenge of jointly estimating the poses, 3D structure, and conformational heterogeneity of a biomolecule from millions of noisy and randomly oriented 2D projections in a computationally efficient manner, however, remains unsolved. Our method, cryoFIRE, performs ab initio heterogeneous reconstruction with unknown poses in an amortized framework, thereby avoiding the computationally expensive step of pose search while enabling the analysis of conformational heterogeneity. Poses and conformation are jointly estimated by an encoder while a physics-based decoder aggregates the images into an implicit neural representation of the conformational space. We show that our method can provide one order of magnitude speedup on datasets containing millions of images, without any loss of accuracy. We validate that the joint estimation of poses and conformations can be amortized over the size of the dataset. For the first time, we prove that an amortized method can extract interpretable dynamic information from experimental datasets.

Keywords: Neural Scene Representation · Cryo-electron microscopy · 3D Reconstruction
[ Paper]  [ Poster]  [ OpenReview] 

Author Information

Axel Levy (Stanford University)
Axel Levy

I am a PhD candidate in Electrical Engineering at Stanford University. I am jointly supervised by Pr. Mike Dunne and Pr. Gordon Wetzstein. My research focuses on the application of inverse rendering methods to scientific imaging (eg. inferring the 3D structure of proteins from cryo-EM images), and more generally in solving inverse problems that arise in physics and biology.

Gordon Wetzstein (Stanford University)
Julien N.P Martel (Stanford University)
Frederic Poitevin (SLAC National Accelerator Laboratory)
Ellen Zhong (Princeton University)

More from the Same Authors