All algorithms
Post-processing
Stage
4
of 8
Priced add-on
LigLoc
Our deep learning model for systematic, unsupervised localisation of ligand density in cryo-EM maps, so every map is assessed the same way instead of by eye.
Trained at scale
trained on over 10,000 maps including ligand, ion and lipid densities.
Unbiased
LigLoc enables unbiased ligand localization, only requiring a 3D density map as input.
What it does
Finding where ligands bind in a cryo-EM map is usually a subjective task. Densities are inspected manually, and two people can reasonably disagree. At the scale of structure-based drug discovery, with many maps and many compounds, this step becomes slow and inconsistent.
LigLoc localises ligands systematically and unsupervised. Every map is assessed the same way, so assignment is reproducible.
How it works
LigLoc uses a deep learning model trained on more than 10,000 ligands, covering lipids, ions and small molecules. You provide a cryo-EM density map and It returns coordinates, a score and a volume estimate for each predicted binding site, so the candidates arrive ranked rather than as a list to triage by eye. LigLoc does not require any prior information on the ligand binding site, no protein model, and no massks to bias the search.
Because the same model looks at every map, results can be compared across a series. Ligloc enables the systematic scoring of hundreds of small molecules in hours, against days of non-systematic, manual work.
Schematic, not measured data. Predicted binding sites ringed on a density map, and the output: coordinates and a score for each site.
Where it sits
LigLoc sits at the very end of the workflow, on a finished map, where the question has moved from structure to binding. It works on the maps the rest of the pipeline produces, after Polishing and post-processing. LigLog supports providing models as an inpute to provide context to localized densities and facilitate interpretation.
It is an add-on to the platform licence, quoted separately.
At a glance
Stage
Post-processing
Runs
On request, on a finished map
Hardware
GPUs
Inputs
A cryo-EM density map
Outputs
Binding site coordinates, scores and volume
One row per predicted site, ranked by score.
Replaces
Availability
Add-on
Assess every map's ligand density the same way.
Put this algorithm to work
See the full system around the workflow step, then choose the right way to get started.
TECHNOLOGY OVERVIEW
Explore all algorithms
Browse the complete analysis workflow, from pre-processing to refinement.
View algorithm overview
PLAN WITH CONFIDENCE
See what’s included
Compare plans, compute, and storage with transparent costs for every workflow.
Explore pricing
SEE YOUR DATA FLOW
Book a guided demo
Talk through your data, workflows, and deployment needs with a CryoCloud specialist.
Book a demo
Next in the workflow
Post-processing
Automated Map Segmentation
Given the number of subunits, a map is segmented into its subunits automatically.
Post-processing
CryoCloud Polishing
Per-particle motion refinement for the last stretch of resolution.
Refinement
CTF Refinement
Fits defocus, astigmatism, beam tilt, aberrations and anisotropic magnification in one job, 3x faster.