2021Nashed CryoPoseNet

From 3DEM-Methods
Jump to navigation Jump to search

Citation

Nashed, Y. S.; Poitevin, F.; Gupta, H.; Woollard, G.; Kagan, M.; Yoon, C. H. & Ratner, D. CryoPoseNet: End-to-End Simultaneous Learning of Single-Particle Orientation and 3D Map Reconstruction From Cryo-Electron Microscopy Data. Proc. IEEE/CVF Intl. Conf. on Computer Vision, 2021, 4066-4076

Abstract

Cryogenic electron microscopy (cryo-EM) provides im-ages from different copies of the same biomolecule in ar-bitrary orientations. Here, we present an end-to-end unsu-pervised approach that learns individual particle orienta-tions directly from cryo-EM data while reconstructing the3D map of the biomolecule following random initialization.The approach relies on an auto-encoder architecture wherethe latent space is explicitly interpreted as orientations usedby the decoder to form an image according to the physi-cal projection model. We evaluate our method on simulateddata and show that it is able to reconstruct 3D particle mapsfrom noisy- and CTF-corrupted 2D projection images of un-known particle orientations.

Keywords

Links

https://openaccess.thecvf.com/content/ICCV2021W/LCI/html/Nashed_CryoPoseNet_End-to-End_Simultaneous_Learning_of_Single-Particle_Orientation_and_3D_Map_ICCVW_2021_paper.html

Related software

Related methods

Comments