Quick summary of cryo-EM

Cryo-electron microscopy (cryo-EM) emerged from advances in electron microscopy developed in the 1930s, but its application to biological macromolecules was historically limited by radiation damage and dehydration artifacts. A major breakthrough occurred in the early 1980s when Jacques Dubochet introduced vitrification of aqueous samples, enabling preservation of biological specimens in a near-native, non-crystalline amorphous ice. Subsequent methodological innovations by Joachim Frank, who developed statistical frameworks for single-particle analysis, and Richard Henderson, who demonstrated that electron microscopy could achieve near-atomic resolution (notably with bacteriorhodopsin), established the foundation of modern cryo-EM. The so-called “resolution revolution” beginning around 2012—driven by the advent of direct electron detectors (DEDs), improved phase plates, and advanced image-processing algorithms—enabled routine high-resolution structure determination and culminated in the 2017 Nobel Prize in Chemistry awarded to these three scientists.

The cryo-EM workflow begins with rapid vitrification of purified biomolecules, typically by plunge-freezing into liquid ethane cooled by liquid nitrogen, producing a thin film of amorphous ice. Samples are then imaged using high-end transmission electron microscopes under low-dose conditions to minimize radiation damage, yielding dose-fractionated movies rather than static micrographs. Subsequent computational processing includes motion correction, contrast transfer function (CTF) estimation, particle picking, 2D classification, ab initio model generation, 3D classification, and high-resolution refinement. These steps produce Coulomb potential maps, into which atomic models are built and refined using real-space refinement approaches.

In contrast to X-ray crystallography, cryo-EM does not require crystallization and is particularly well suited for membrane proteins, large macromolecular assemblies, and structurally heterogeneous or flexible systems. Importantly, cryo-EM can resolve multiple conformational states from a single dataset through classification strategies, providing direct insights into functional dynamics. In drug discovery, cryo-EM enables visualization of ligand–target interactions at near-atomic resolution, supporting structure-based drug design (SBDD) and fragment screening approaches. During the COVID-19 pandemic, cryo-EM facilitated rapid determination of the SARS-CoV-2 spike protein structure, significantly accelerating vaccine and therapeutic development.

Ongoing methodological developments include automated and reproducible sample preparation (e.g., vitrification robotics and microfluidics), AI-driven image processing for particle picking and heterogeneity analysis, and time-resolved cryo-EM approaches aimed at capturing transient intermediates. Cryo-electron tomography (cryo-ET), particularly when combined with subtomogram averaging, extends structural analysis into native cellular environments. Despite persistent challenges—such as high instrumentation and maintenance costs, beam-induced motion, preferred orientation, and the need for specialized expertise—cryo-EM has become a central technique in structural biology. As innovations continue, cryo-EM is poised to further advance mechanistic biology and accelerate rational therapeutic development.

Study material for Cryo-EM basics

 Tutorial Playlists 

 Online resources 

 Data processing software we use