Observed Antibody Space (OAS)
Repository of >1B antibody BCR sequences from public repertoires — core for antibody LM pretraining.
Composite
97.5
Experimental validation
Retrospective
Stages
Hit IDLead ID / ADMET
Modalities
biologic-mab
Task types
antibody-sequence
Size
sequences: 2,400,000,000
repertoires: 15,000
repertoires: 15,000
License
CC-BY 4.0
First release
2018
Last updated
2025-04
Official site
Leaderboard
→ leaderboard
Dataset
Code / GitHub
→ repository
HuggingFace
→ HF
Paper
Observed Antibody Space: A diverse database of cleaned, annotated, and translated unpaired and paired antibody sequences · Olsen TH, Boyles F, Deane CM · 2022 · paper · doi:10.1002/pro.4205 · 310 citations
Flags
none
Experts
Groups
Hosted by
—
Related benchmarks
Rubric (7-criterion)
rigor
5
coverage
5
maintenance
5
adoption
5
quality
4
accessibility
5
industry_relevance
5
Notes
Underlies AbLang, IgLM, AntiBERTa — industry-adopted.