Smart Parallel Automated Cryo Electron tomography

Abstract

In situ cryo electron tomography enables investigation of macromolecules in their native cellular environment. Data collection, however, requires an experienced operator and valuable microscope time to carefully select targets for tilt series acquisition. Here, we developed a workflow using machine learning approaches to fully automate the entire process, including lamella detection, biological feature segmentation, target selection, and tilt series acquisition, all without the need for human intervention.