Use LocScale Online

LocScale: Local Sharpening of Cryo-EM Density Maps Using an Atomic Reference Model

LocScale is a procedure for local sharpening of cryo-EM density maps based on prior knowledge of an atomic reference structure, introduced by Arjen J. Jakobi, Matthias Wilmanns and Carsten Sachse in the eLife paper "Model-based local density sharpening of cryo-EM maps" (2017). It improves the interpretability of density maps so that model building and refinement become easier. The current LocScale-2.0 repository describes an automated map optimisation program that performs physics-informed local sharpening and/or density modification.

Atomic models built into high-resolution density maps are the ultimate result of the cryo-EM structure determination process. Real maps, however, often vary in resolution and contrast from one region of the structure to another, so a single global sharpening setting cannot serve every region equally well. LocScale addresses this by sharpening each local neighbourhood against a reference derived from an atomic model.

How LocScale Sharpens Cryo-EM Maps Locally

The procedure optimizes the contrast of cryo-EM densities by amplitude scaling against the radially averaged local falloff estimated from a windowed reference model. In practice this means:

1. Reference model: an atomic structure supplies prior knowledge of how density amplitudes should fall off with resolution.

2. Windowing: the map and the reference are compared within local windows rather than globally.

3. Scaling: the local amplitudes of the experimental map are scaled to match the reference falloff, increasing the contrast of the density.

Modes in the current software

• Model-based (legacy) sharpening uses an atomic model as the reference.

• Hybrid sharpening performs reference-based local sharpening with partial models.

• Model-free sharpening works without atomic models.

• Feature-enhanced maps use confidence-weighted optimisation by variational inference.

Results: Better Interpretability Across Six Cryo-EM Structures

The authors tested the procedure on six cryo-EM structures and showed that local sharpening can increase the interpretability of density maps, in particular where resolution varies across the map, and that it facilitates model building and atomic model refinement.

Is the Sharpening Biased by the Reference Model?

The authors checked this directly. Side-chain densities remained readily interpretable even when the reference model contained no side-chain information, and when atom positions were randomly perturbed by 2.5 to 50 Å r.m.s.d., although map contrast decreased as the perturbation grew. They conclude that the observed features result from enhanced contrast rather than coordinate bias from the reference.

What is Tamarind Bio?

Tamarind Bio is a no-code bioinformatics platform built to give life scientists and researchers access to powerful computational tools. Many cutting-edge machine learning models are hard to deploy and use. Tamarind provides an intuitive, web-based environment that removes the complexity of high-performance computing, software dependencies and command-line interfaces.

The platform is designed for biologists, chemists and other researchers who may not have a background in programming or cloud infrastructure but want to run models on their own data. Key features include:

• A user-friendly graphical interface for setting up and launching experiments

• A robust API for integration into existing research pipelines

• An automated system for managing and scaling computational resources

Tamarind treats information and data security as a top priority, as detailed in its Trust Center and Terms of Service.

Accelerating Discovery with LocScale on Tamarind Bio

• Model building: sharper local density makes it easier to trace and build atomic models.

• Model refinement: improved contrast supports refinement of existing models.

• Heterogeneous-resolution maps: the authors highlight cases where resolution varies across the map.

• Map interpretation: better local contrast helps interpret flexible or lower-resolution regions.

How to Use LocScale on Tamarind Bio

1. Open the tool: log in to tamarind.bio and select LocScale.

2. Provide the density map: upload your cryo-EM map.

3. Provide the reference model: for model-based sharpening, upload the atomic model fitted to the map.

4. Set parameters: adjust options such as the local window size, if exposed.

5. Run the job: submit the job and wait for it to finish.

6. Download the outputs: retrieve the sharpened map.

7. Interpret and use: inspect the sharpened map in a viewer and use it for model building or refinement.

Things to Keep in Mind

• The original procedure depends on prior knowledge of an atomic reference structure, so reference quality matters.

• For best performance the method needs good estimates of atomic B-factors for the reference, although the authors report it still improves maps when B-factors are not yet refined.

• Window size is a real parameter: windows of 14 pixels and smaller attenuate low frequencies, and the smallest practical window is eight pixels.

• Model-map FSC and real-space correlation comparisons are only meaningful against a common reference map that has not undergone model-based scaling.

• Model-free and hybrid modes exist in LocScale-2.0 for cases without a complete model.

• The source pages do not specify input file formats, so check the Tamarind tool form for accepted formats.

Source: Jakobi AJ, Wilmanns M, Sachse C. Model-based local density sharpening of cryo-EM maps. eLife 6:e27131 (2017). DOI: 10.7554/eLife.27131. Code: https://github.com/cryoTUD/locscale.