# PRISM Antibody Training Data: 66 Million Unpaired and 764k Paired Human Antibody Sequences from OAS with Germline Labels (Romero Lab, Duke)

## Facts

- Provider: Romero Lab, Duke University
- License: CC-BY-4.0 (https://huggingface.co/datasets/RomeroLab-Duke/prism-antibody-data)
- Size: 5 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/RomeroLab-Duke/prism-antibody-data
- Access: public, readable without an account
- Category: Biology
- Source: https://opig.stats.ox.ac.uk/webapps/oas/
- Page: https://scigantic.com/archive/631526b8-edff-5de6-9103-c0d91e1aec08
- Keywords: Biology, Parquet, Hugging Face

## About

Curated human antibody sequence data used to train PRISM, a germline-aware antibody language model from the Romero Lab at Duke University. Sequences come from the Observed Antibody Space (OAS), human studies only (70 studies, 2021 to 2022 data freeze), passed through quality filters (missing conserved cysteine, heavy fragmentation, non-canonical amino acids), exact deduplication, a somatic mutation filter (more than 3 non-germline mutations in heavy and 2 in light chains), CDR3 and whole-sequence clustering, and ANARCI re-numbering. Two Parquet files hold 66,435,576 unpaired heavy and light chains for pretraining and 763,989 paired VH and VL antibodies for fine-tuning, with amino acid and nucleotide sequences, germline alignments, B-cell type, somatic mutation codes and counts, V and J gene calls, IMGT region masks, cluster IDs and a train/valid/test split column. The OAS download manifest is included; most of the Briney 2019 heavy-chain study was left out for size. Use it for antibody language models, somatic hypermutation and B-cell repertoire analysis. Derived from OAS under CC BY 4.0; cite Olsen et al. 2022, Protein Science, and PRISM.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("RomeroLab-Duke/prism-antibody-data")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `paired_anarci_relabeled.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/631526b8-edff-5de6-9103-c0d91e1aec08
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/631526b8-edff-5de6-9103-c0d91e1aec08
- Over MCP: `get_schema_card` with id `631526b8-edff-5de6-9103-c0d91e1aec08` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Biology datasets](https://scigantic.com/topics/biology)
