CryoEM single particle analysis (SPA) uses the electron scattering mechanism combined with cryogenic freezing, which significantly reduces high-energy electron beam damage to molecular structures, enabling near-native high-resolution structure determination of proteins without crystallization.
A systematic eight-stage quality-controlled pipeline covering every step from sample receipt through PDB deposition and final report delivery.
In cryo-EM SPA, tens of thousands of 2D projection images of individual protein particles are collected, computationally aligned, and averaged to reconstruct a high-resolution 3D density map. Unlike crystallography, no ordered crystal is needed — particles are imaged in their near-native hydrated state embedded in vitrified ice.
The process begins with negative-stain TEM for rapid quality assessment, followed by cryo-sample preparation and high-resolution data collection on a 300 kV FEG instrument. Data processing produces density maps with resolution assessed by gold-standard Fourier shell correlation, into which atomic models are then built and refined.
A key advantage is the ability to capture multiple conformational states within a single experiment — 3D classification separates heterogeneous particle populations into distinct structural states, providing a dynamic view of protein function that is inaccessible to crystallography.
The client provides purified protein samples directly, or specifies the target protein sequence for ICDMO to arrange expression and purification through partner laboratories. Sample purity, concentration, and buffer conditions are assessed and confirmed before proceeding to the next stage.
Heavy metal staining (uranyl acetate or uranyl formate) is applied for rapid quality assessment. Micrographs are analyzed to evaluate particle homogeneity, dispersity, size, shape, and sample concentration — providing a fast go/no-go decision before cryo-sample preparation.
Using a Vitrobot Mark IV plunge-freezer, 3–5 µL of protein sample is applied to a glow-discharged grid. The sample is blotted to form a thin molecular layer and rapidly plunged into liquid ethane (−185°C), vitrifying the sample in amorphous ice to preserve the native protein structure and spatial distribution.
Grids are screened using a 200 kV FEG microscope to assess ice thickness, particle distribution, preferred orientation, and 2D class average quality. Preliminary data collection provides resolution estimates and guides optimization of grid conditions before high-resolution acquisition.
High-resolution data is collected on a 300 kV FEG cryo-electron microscope (Titan Krios) equipped with a direct electron detector (Falcon 4 or K3). Automated acquisition software (EPU, SerialEM) collects tens of thousands to millions of single-particle micrographs with high-throughput automation.
Raw micrograph movies are processed through a multi-step pipeline: beam-induced motion correction, contrast transfer function estimation, automated particle picking, two- and three-dimensional classification, ab initio reconstruction and iterative refinement. Resolution is tracked by gold-standard Fourier shell correlation throughout, and terabyte-scale datasets are handled on dedicated compute.
The refined density map is used for atomic model building, with predicted or homologous starting models fitted into density and rebuilt where the map supports it, followed by real-space refinement. Model geometry and fit to density are validated against deposition-standard criteria, and regions not supported by the map are reported as unmodelled rather than filled in.
The final atomic structure is analyzed for biological insights: binding interfaces, conformational changes, potential drug-binding pockets, and mechanistic interpretation. A comprehensive project report is delivered including the density map, atomic coordinates, local resolution maps, Fourier shell correlation curves and publication-quality figures.
MRGPRX2 (Mas-Related G Protein-Coupled Receptor X2) is a GPCR primarily expressed in skin mast cells. It is mainly involved in non-IgE-dependent mast cell activation, playing a key role in Type VII hypersensitivity reactions and pseudo-allergic drug responses.
Using cryo-EM single particle analysis, ICDMO determined the near-atomic resolution structure of the MRGPRX2 complex with its G protein partner. The structure revealed the transmembrane helix packing geometry, key ligand-binding residues in the orthosteric pocket, and a cryptic allosteric site — providing the molecular basis for mast cell activation and a structural template for drug safety profiling and therapeutic development.
Note: Timelines are estimates for well-behaved samples. Challenging membrane proteins or heterogeneous complexes may require additional optimization. Contact us for a free negative stain pre-screening before committing to full cryo-EM data collection.
Submit your sample for a free negative stain TEM screening. We assess particle quality and provide a resolution feasibility estimate before you commit to full cryo-EM data collection.
Contact Us Online ConsultationMap quality sets a hard ceiling on interpretation, and no processing recovers information the experiment did not capture. Preferred orientation, air-water interface damage and sample heterogeneity limit achievable resolution and are properties of the sample rather than the software. Regions with weak density are reported as unmodelled instead of built speculatively. Sample preparation, grid screening and data collection are performed by your laboratory or a partner facility.
The results below were published by independent groups and are cited to show what these computational approaches have achieved in the literature. They are not ICDMO project results, and outcomes vary substantially between targets, datasets and programmes. We do not promise comparable results for any particular project.
Current deep-learning structure prediction, evaluated on an independent protein–ligand benchmark, placed roughly 76% of ligands within 2 Å of the experimental pose, and reported approximately 80% by coordinate error — a substantial improvement over classical docking baselines on the same set, achieved without supplying a binding site. The corollary matters as much as the headline: roughly a quarter of cases still fall outside that threshold, which is why we report confidence per interface rather than delivering a single model.
Abramson et al., Nature, 2024 ↗