How Does Cryo-EM Work? A Beginner's Guide
Plain-language explanation of cryo-electron microscopy — how samples are frozen, imaged, and reconstructed into 3D molecular structures.
The Basic Principle
Cryo-electron microscopy (cryo-EM) determines the three-dimensional structures of biological molecules — proteins, protein complexes, and viruses — by imaging them with an electron beam while they are frozen in a thin layer of vitreous (glass-like) ice. Unlike X-ray crystallography, cryo-EM does not require crystals, making it possible to study molecules that are difficult or impossible to crystallize, including large, flexible, and membrane-bound complexes.
Step 1: Sample Vitrification
The process begins by applying a tiny drop (3–5 µL) of purified protein solution to a metal grid coated with a thin carbon film containing small holes. The grid is then blotted with filter paper to create a thin aqueous layer spanning the holes, and immediately plunged into liquid ethane cooled by liquid nitrogen. This rapid freezing (in milliseconds) traps the water as vitreous ice rather than crystalline ice, preserving the protein molecules in their native, hydrated state. The most common plunge-freezing instrument is the Vitrobot Mark IV.
Step 2: Electron Imaging
The frozen grid is loaded into a cryo-transmission electron microscope (cryo-TEM) like the Titan Krios G4, which operates at 300,000 volts and maintains the sample below −170°C. The electron beam passes through the thin ice layer, and a direct electron detector (such as the Falcon 4i or Gatan K3) captures images — called micrographs — of hundreds of individual protein molecules embedded in the ice. Each molecule is oriented randomly, providing different views of the same structure. Because biological molecules are very sensitive to radiation damage, the total electron dose must be carefully controlled.
Step 3: Single Particle Reconstruction
Software (typically cryoSPARC or RELION) identifies and extracts hundreds of thousands to millions of individual particle images from the micrographs. These 2D images are computationally classified and averaged to improve signal-to-noise, then used to reconstruct a 3D density map of the molecule. The best structures now routinely reach 2–3 Å resolution, sufficient to visualize individual amino acid side chains and bound drug molecules.
Why Cryo-EM Matters
Cryo-EM has revolutionized structural biology since the mid-2010s (the 'resolution revolution'), earning the 2017 Nobel Prize in Chemistry. It is now the method of choice for solving structures of large protein complexes, membrane proteins, ribosomes, and virus particles. In drug discovery, cryo-EM enables structure-based drug design for targets that were previously intractable by X-ray crystallography. The technique continues to advance rapidly, with new instruments, detectors, and software pushing resolution boundaries.
Frequently Asked Questions
What resolution can cryo-EM achieve?
The best cryo-EM structures reach 1.2–1.5 Å resolution for favorable samples (e.g., apoferritin). For typical drug targets and protein complexes, 2–4 Å is common. Resolution depends on particle size, flexibility, ice quality, and data quantity.
How long does a cryo-EM experiment take?
Sample preparation takes 1–2 days (including grid optimization). Data collection on a Titan Krios runs 24–72 hours per dataset. Data processing takes 1–7 days depending on dataset size and computational resources. A full project from purified protein to 3D map typically takes 2–6 weeks.
What is the minimum protein size for cryo-EM?
Conventional single particle cryo-EM works best for molecules larger than ~50–100 kDa. Below that size, particles are difficult to distinguish from noise in the ice. However, techniques like scaffolding (attaching small targets to larger carriers), Fab/nanobody complexes, and advances in energy-filtered imaging are pushing the lower limit below 50 kDa.
Related Resources
Need Help Finding a Facility?
Browse our directory or get a fast-track quote from a private CRO.