Learning the Shared Latent Structure of Cryo-EM Inverse Problems
The EMber family of vision transformers are introduced, pre-trained on>50 million experimental particle images from 214 EMPIAR datasets together with simulated projections from 268,000 molecular assemblies, and it is demonstrated how to learn representations that capture biological information for use across multiple c...
N. Chalapathi, Theodore Sternlieb, Ethan Han et al.
· 0 citations