BioBench Atlas
Benchmarks/ProteinGym/ProteinGym Clinical Indels

track · audited · verified 2026-07-21

ProteinGym Clinical Indels

ProteinGym track for classifying short human clinical insertion and deletion variants against ClinVar and gnomAD-derived labels.

Protein sequenceClinical and translationalProtein clinical variant interpretation

Benchmark definition

What is counted

Version
1.3
Total
1555 (clinical proteins)
Task formats
clinical variant classification
Capabilities
PredictionClassification
Modalities
Protein sequenceTable

Version history

VersionStatusRelease / as-ofTotalFormal tracks
1.0
proteingym-clinical-indel-v10		superseded2023-12-081555 (clinical proteins)None registered
1.3
proteingym-clinical-indel-v13		current2025-04-271555 (clinical proteins)None registered

Scientific Task Atlas

Scientific task classification

complete for 1.3. Formal clinical indel track.

Scientific taskCoverageCountMappingEvidence
Protein clinical variant interpretationexplicitly-in-scope1555 records
Clinical proteins.		official-track
high confidence		proteingym-clinical-indel-evidence-v13
The registry count unit is proteins, not variants.

Evaluation registry

Works and run settings

A setting change—scope, prompt, tools, budget, grader, or repeats—creates a separate run. Charts never cross a comparability group.

No normalized evaluation run is published yet. Creator evidence is still attached below.

Evidence and change history

Source locators remain visible; expand an item to inspect the exact Registry fields it supports.

ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · section: pp. 1–10 and 31, Sections 3–5 and Appendix A.3.2 (Track definition, identity, domains, modalities, classification format, metric, and access.) · Supports 10 fields

Open source →

  • /name
  • /organizations
  • /release_date
  • /latest_version
  • /kind
  • /summary
  • /domains
  • /capabilities
  • /modalities
  • /task_formats
ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · table: pp. 6 and 29, Tables 1 and A1 (1,555 clinical indel proteins in v1.0.) · Supports 3 fields

Open source →

  • /versions/0/task_counts/total
  • /versions/0/task_counts/basis
  • /versions/0/task_counts/subsets
proteingym-clinical-indel-v13-resource · repository-path: clinical_indels.csv in Zenodo record 15293562 (1,555 reference rows, one per clinical protein.) · Supports 7 fields

Open source →

  • /task_counts/total
  • /task_counts/basis
  • /task_counts/subsets
  • /versions/1/task_counts/total
  • /versions/1/task_counts/basis
  • /versions/1/task_counts/subsets
  • /scientific_task_classification/entries/0
proteingym-clinical-indel-repository-resource · repository-path: README.md Overview, Results, Resources, Usage and reproducibility, and License (Current version, downloads, scoring, implementation, and MIT license.) · Supports 9 fields

Open source →

  • /latest_version
  • /access/level
  • /access/tasks
  • /access/artifacts
  • /access/grader
  • /access/license
  • /access/biosafety_notes
  • /resources
  • /implementations

View source-level modification history on GitHub →