ProteinGym
217 DMS substitution assays + indel + clinical variants — de facto standard for VEPs.
Composite score: 97.5
Rubric (1–5 per criterion)
rigor
5/5
coverage
5/5
maintenance
5/5
adoption
5/5
quality
5/5
accessibility
5/5
industry_relevance
4/5
Metadata
Stages
Target IDLead ID / ADMETIND-enabling
Modalities
protein-general
Task types
variant-effect
License
MIT
First release
2022
Last updated
2025-03
Flags
none
Size & scope
- dms_assays: 217
- mutations: 2700000
- clinical_variants: 2525
Primary paper
Title
ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design
Authors
Notin P, Kollasch A, Ritter D, et al.
Year
2023
DOI / arXiv
Citations
320
Links
- Official site: https://proteingym.org/
- GitHub: https://github.com/OATML-Markslab/ProteinGym
- Leaderboard: https://proteingym.org/benchmarks
Hosted by (initiatives)
Experts (primary authors / maintainers)
Groups (host labs / companies / consortia)
Related benchmarks
Notes (honest caveats)
Field standard. Clinical track enables fair ESM/EVE/AlphaMissense comparison.