Cryo-electron microscopy (cryo-EM) has become a major imaging modality for atomic resolution structural biology. Early optimism around cryo-EM ability to resolve the atomic structures of nanometer macromolecular machines that are not amenable to other imaging techniques. Furthermore, the related technique of cryo-electron tomography (cryo-ET) has also become indispensable for examining the organization and macromolecular structure of cells at nanometer resolution. The far-reaching prospect, however, comes from the possibility of both techniques to resolve the structure and dynamics of macromolecular machines to near-atomic resolution.
Currently the cryo-EM community faces unprecedented data loads where the only practical way to analyze them requires a heavy reliance on specialized processing algorithms. Recent breakthroughs in machine learning have created opportunities to augment these algorithms with human intuition, complex prior knowledge, and robust validation schemes.
This unique IMS-sponsored workshop gathers international specialists in cryo-EM, machine learning, and mathematical sciences to share and inspire future applications of machine learning in cryo-EM.