Master the EWCL Analysis Dashboard in minutes
EWCL (Entropy-Weighted Collapse Likelihood) quantifies residue-level collapse risk using entropy, curvature, and hydropathy cues. This guide walks through loading structures, reviewing outputs, and exporting artifacts for your workflows.
Supported inputs
PDB files, pasted text, PDB IDs, UniProt IDs, EWCL demo bundle
Runtime tips
Average run time: 5-20s per structure. Use reverse EWCL for IDPs.
Upload a PDB file
Drag and drop or browse for any .pdb or .ent file. Larger files are supported – we stream and parse them efficiently.
Paste raw PDB text
Switch to the Paste tab and drop in the contents of a PDB file. Perfect for quick checks from notebooks or scripts.
Fetch by PDB or UniProt ID
Enter a 4-character PDB ID (e.g. 1UBQ) or supported UniProt accession. We’ll retrieve, parse, and queue the structure automatically.
Run an analysis
Quick checklist before you hit “Run EWCL analysis”.
- 1. Choose a mode. Standard EWCL emphasizes collapse risk for folded domains; Reverse EWCL highlights disorder for IDPs.
- 2. Validate metadata. Chains, residue counts, and organism labels are surfaced immediately so you can confirm the structure matches expectations.
- 3. Queue the analysis. The dashboard streams progress and prepares visualizations, tables, and exports as soon as scores are computed.
Pro tip
Switch to demo proteins from the sidebar for curated examples that showcase EWCL vs DisProt correlations and hallucination detection.
Interpret your results
Everything the dashboard surfaces after a successful run.
3D EWCL Viewer
Residues are colored by collapse likelihood. Click any residue to sync selection with the analysis tables.
Residue & Feature Tables
Sort, search, and export tables that include EWCL, reverse EWCL, B-factor, pLDDT, and hallucination metrics.
Distribution & Heatmaps
Score distributions, donut charts, and entropy heatmaps highlight stability classes and cryptic regions at a glance.
Correlation Insights
Pearson/Spearman correlations versus B-factors or pLDDT let you benchmark EWCL against canonical structure confidence.
Click any residue in the 3D viewer to auto-scroll to the corresponding row in the Residue Analysis table—ideal for reviewing cryptic regions.
Export & share
Get results into downstream notebooks, reports, or pipelines.
Use the export controls above the results cards. JSON captures the complete dataset (residues, metadata, metrics), while CSV offers a tidy table for quick filtering inside R or pandas.