UCLA researchers have developed an automated Microcrystal Electron Diffraction (MicroED) microscopy system and method that autonomously identifies, images, and analyzes micro-crystals on an EM grid. The system captures low-magnification atlases, selects promising grid squares, acquires higher-magnification montages, sets focus, captures continuous-rotation MicroED videos, estimates crystal dimensions, determines symmetry or unit cell parameters, and outputs compositional analysis of compound mixtures.
Structural analysis of micro-crystalline materials (organic, inorganic, biological) via electron diffraction is valuable for determining atomic structures, especially when crystals are too small for X-ray crystallography. However, current workflows require considerable manual oversight: choosing grid squares, adjusting focus, identifying suitable crystals, setting data collection parameters, and performing analyses one crystal at a time. These manual steps are time-consuming, subject to human error, low throughput, and inefficient when dealing with mixtures of compounds or large numbers of crystals.
This technology integrates microscopy automation, image analysis, and computational crystallography in a MicroED setup. The method includes automatically capturing a low-magnification atlas of a grid with mixed compound crystals, algorithmically selecting grid squares based on crystal count or quality metrics, moving to medium magnification to create montages, setting eucentric focus, recording continuous rotation MicroED videos of selected crystals, estimating crystal physical dimensions, computing unit-cell dimensions and symmetry, and then producing compositional analysis of the mixture based on those structural features. The system is designed to run with minimal human input, enabling high-throughput analysis of microcrystals.
Automation reduces manual labor and human error in MicroED workflows.
Increased throughput—can process many crystals across mixtures without needing manual screening.
Enables compositional analysis of compound mixtures, not just single pure crystals.
Focused imaging (medium magnification) guided by earlier maps saves time.
Estimation of crystal dimensions and symmetry helps discard unsuitable crystals early.
Provides structural metrics (unit cell, symmetry) that feed downstream crystallography or material identification.
Structural determination of novel small molecules, proteins, organic crystals, or pharmaceuticals.
Material science for identifying phases in mixed powders or nanocrystals.
Drug discovery / pharmaceutical companies needing fast structure confirmation.
High-throughput characterization in crystallography and chemoinformatics.
Automated screening facilities or core labs with many samples to process.
WO2024/233601 A2 — Systems and Methods for Autonomous MicroED Microscopy Systems and Methods for Autonomous MicroED Microscopy (WO2024233601A2) Google Patents