PEER
PEER — 14 protein property prediction tasks across structure/function/interaction.
Composite score: 74.4
Rubric (1–5 per criterion)
rigor
4/5
coverage
4/5
maintenance
3/5
adoption
3/5
quality
4/5
accessibility
5/5
industry_relevance
3/5
Metadata
Stages
Target IDDevelopmental Candidate
Modalities
protein-general
Task types
protein-ml
License
MIT
First release
2022
Last updated
2024
Flags
none
Size & scope
- tasks: 14
Primary paper
Title
PEER: A Comprehensive and Multi-Task Benchmark for Protein Sequence Understanding
Authors
Xu M, Yuan X, Miret S, Tang J
Year
2022
DOI / arXiv
Citations
130
Links
- Official site: https://github.com/DeepGraphLearning/PEER_Benchmark
- GitHub: https://github.com/DeepGraphLearning/PEER_Benchmark
- Leaderboard: N/A
Hosted by (initiatives)
Experts (primary authors / maintainers)
Groups (host labs / companies / consortia)
Related benchmarks
Notes (honest caveats)
Broader than TAPE, tighter than ProteinGym; good middle ground.