Glossary

Local Resolution Estimation

A computational method that calculates a resolution value for each voxel in a cryo-EM density map, identifying regions of structural flexibility or disorder.

Get in touch Learn more

Moody home-office setup in a converted highrise loft, analyst working late with multiple screens showing knowledge graph visualizations, city lights through large windows behind.

Voxel-Wise Map Quality

What is Local Resolution Estimation?

A computational method that calculates a resolution value for each voxel in a cryo-EM density map, identifying regions of structural flexibility or disorder.

Local Resolution Estimation is a computational method that assigns a distinct resolution value to every voxel within a cryo-EM density map, moving beyond a single global metric to reveal the map's spatially varying quality. This technique identifies rigid, well-ordered domains with high resolution and flexible, disordered regions with lower resolution, providing a quantitative map of structural heterogeneity.

Algorithms typically compute local resolution by calculating the Fourier Shell Correlation (FSC) within a sliding window or around individual atoms, producing a color-coded heatmap for intuitive visualization. Tools like ResMap and MonoRes implement this analysis, enabling researchers to validate atomic models, interpret flexible loops, and understand the confidence of structural features in different map regions.

Voxel-Wise Quality Assessment

Key Characteristics of Local Resolution Estimation

Local resolution estimation moves beyond a single global number to map the spatially varying quality of a cryo-EM density map, revealing flexible domains, disordered loops, and the impact of preferred orientation.

Voxel-Based Resolution Mapping

Calculates a resolution value for every voxel in a 3D cryo-EM map, producing a color-coded heatmap. This identifies rigid, well-resolved core regions (often <3 Å) versus flexible, poorly resolved peripheral domains (>5 Å).

Input: Two independently refined half-maps from gold-standard refinement
Output: A 3D volume where voxel intensity encodes local resolution in Ångströms
Key Insight: A map with a 'global' resolution of 3.2 Å may contain a catalytic site at 2.8 Å and a mobile loop at 6 Å

Per-voxel

Granularity

Fourier Shell Correlation (FSC) in a Sliding Window

The core algorithm computes the Fourier Shell Correlation (FSC) between two half-maps within a small, moving window centered on each voxel. The resolution is defined as the spatial frequency where the FSC drops below a threshold (commonly 0.143 or 0.5).

Window size is a critical parameter: too small introduces noise; too large smooths out genuine variation
Gold-standard FSC principles are applied locally to prevent overfitting
Implemented in tools like ResMap, MonoRes, and cryoSPARC's Local Resolution

0.143

Gold-standard FSC threshold

Identifying Conformational Heterogeneity

Local resolution maps are a primary diagnostic for compositional and conformational heterogeneity. Regions with systematically lower local resolution often correspond to:

Flexible loops and termini that adopt multiple conformations
Domain motions where a subdomain moves relative to the core
Partially occupied ligands or binding partners
Glycosylation sites and other flexible post-translational modifications

This guides subsequent 3D variability analysis or focused classification to separate discrete states.

Anisotropy and Preferred Orientation Artifacts

Local resolution estimation reveals directional resolution anisotropy caused by preferred particle orientation at the air-water interface. A map may resolve α-helices clearly in the x-y plane but be smeared along z.

Diagnostic: A 'striped' or directional gradient in the local resolution map
Mitigation: Tilted data collection, stage tilting, or using anisotropic sharpening during post-processing
3D FSC provides a complementary, directionally explicit resolution assessment

Post-Processing and Map Sharpening Guidance

Local resolution estimates directly inform local sharpening algorithms. Instead of applying a single global B-factor, tools like LocalDeblur or DeepEMhancer use the local resolution map to apply spatially varying sharpening.

Sharpening: Amplifies high-frequency Fourier components to enhance atomic detail
Local sharpening prevents over-sharpening of already well-resolved regions and under-sharpening of flexible domains
Results in maps where side-chain detail is visible across a wider range of local quality

Validation and Model Building

Local resolution maps are essential for atomic model validation. They define the local confidence for placing atoms and guide the restraint weighting during real-space refinement.

Model geometry restraints can be relaxed in high-resolution regions and tightened where the map is poor
Model-vs-map FSC and Q-score provide per-residue validation metrics that correlate with local resolution
Prevents over-interpretation: a model should not be built into density at a resolution where side chains are not justified

LOCAL RESOLUTION ESTIMATION

Frequently Asked Questions

Answers to common questions about how local resolution estimation quantifies structural heterogeneity in cryo-EM density maps, enabling researchers to identify flexible domains and validate atomic model interpretation.

Local resolution estimation is a computational method that calculates a resolution value for each voxel in a cryo-EM density map, producing a 3D resolution gradient rather than a single global figure. Unlike gold-standard Fourier shell correlation (FSC), which reports a single resolution for the entire reconstruction, local resolution reveals how map quality varies spatially due to structural flexibility, compositional heterogeneity, or preferred orientation artifacts. The method typically operates by computing FSC curves within small, overlapping sub-volumes or by analyzing local Fourier amplitude correlations. Tools like ResMap, MonoRes, and BLOCRES implement statistical approaches to estimate local resolution without requiring half-map splitting, while cryoSPARC's Local Resolution Estimation uses a windowed FSC approach. The output is a color-coded map where rigid core regions often achieve near-atomic resolution (2-3 Å), while flexible loops or domains may show significantly lower resolution (5-10 Å), providing critical validation for atomic model building and interpretation.

RESOLUTION METRICS

Local vs. Global Resolution Estimation Comparison

A comparison of local and global resolution estimation methodologies for cryo-EM density maps, highlighting their diagnostic scope, output granularity, and computational requirements.

Feature	Global Resolution (FSC)	Local Resolution Estimation	3D Variability Analysis (3DVA)
Primary Metric	Gold-Standard Fourier Shell Correlation (FSC)	Per-voxel FSC or Local FSC	Principal Component Analysis of variance
Output Granularity	Single scalar value (e.g., 3.2 Å)	3D map with a resolution value per voxel	3D map of conformational variance
Identifies Flexible Regions
Requires Half-Maps
Models Continuous Motion
Computational Cost	Low	Moderate	High
Typical Software	RELION, cryoSPARC	MonoRes, ResMap, blocres	cryoSPARC (3DVA)
Diagnostic Use Case	Overall map quality assessment	Identifying disordered domains	Resolving continuous conformational landscapes

Enabling Efficiency, Speed & Accuracy

Intelligent Analysis, Decision & Execution

We build AI systems for teams that need search across company data, workflow automation across tools, or AI features inside products and internal software.

Talk to Us

Search across company data

Give teams answers from docs, tickets, runbooks, and product data with sources and permissions.

Useful when people spend too long searching or get different answers from different systems.

Enterprise searchRAGPermissions

Automate internal workflows

Use AI to route work, draft outputs, trigger actions, and keep approvals and logs in place.

Useful when repetitive work moves across multiple tools and teams.

AI agentsWorkflow automationGovernance

Add AI to products and internal tools

Build assistants, guided actions, or decision support into the software your team or customers already use.

Useful when AI needs to be part of the product, not a separate tool.

AI integrationDecision supportModel routing

CRYO-EM RESOLUTION & VALIDATION

Related Terms

Concepts essential for understanding how local resolution estimation fits into the cryo-EM structure determination pipeline.

Gold-Standard Fourier Shell Correlation (FSC)

The foundational global resolution metric that local estimation extends. The dataset is split into two independent half-sets which are reconstructed separately. The correlation between their Fourier shells is plotted against spatial frequency, with the 0.143 cutoff defining the nominal resolution. This prevents overfitting and noise correlation, but provides only a single value for the entire map, masking local variability.

0.143

FSC Cutoff Criterion

3D Variability Analysis (3DVA)

A method in cryoSPARC that uses principal component analysis to model a continuous landscape of conformational motions. While local resolution estimation identifies where resolution varies, 3DVA reveals how the structure moves. The output is a series of eigenvolumes that can be traversed to visualize molecular breathing, hinge motions, or subunit rotations.

Map Sharpening & B-Factor Weighting

A post-processing step that applies a negative B-factor to Fourier amplitudes to restore high-frequency detail attenuated during imaging. Local resolution estimation directly informs local sharpening—applying different B-factors to different regions based on their estimated resolution. This prevents over-sharpening of flexible domains while enhancing well-resolved cores.

CryoDRGN

A deep generative model using a variational autoencoder to reconstruct continuous conformational heterogeneity. Unlike local resolution estimation which quantifies map quality, CryoDRGN learns a latent space of structural states directly from particle images. The two are complementary: local resolution maps can validate whether CryoDRGN's reconstructed states maintain consistent resolution across the conformational landscape.

Preferred Orientation & Anisotropic Resolution

A common sample preparation artifact where macromolecules adopt limited orientations at the air-water interface. This causes anisotropic resolution—the map is better resolved in some directions than others. Local resolution estimation reveals this as directional variation, often visualized using 3D FSC plots showing resolution along orthogonal axes.

ModelAngelo & Atomic Model Validation

An automated model building tool using a graph neural network to trace the protein backbone into cryo-EM density. Local resolution maps guide model building by indicating which regions support confident side-chain placement versus only backbone tracing. Regions below ~3.5 Å local resolution typically only permit polyalanine-level interpretation.

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.

Limited slotsGet a Free AI Consultation

How We Work

Custom AI workflows for your Business

One-fit-all AI don't work for modern businesses. At Inferensys, we aim to understand your business & custom requirements; which we use to define most efficient agentic workflows, the data, and the tools for your business.