metadata
license: apache-2.0
tags:
- biology
- genomics
- dna
- foundation-models
- arc-institute
- evo2
- science
- huggingscience
pretty_name: OpenGenome2
task_categories:
- text-generation
- fill-mask
- feature-extraction
size_categories:
- n>1T
dataset_info:
features:
- name: text
dtype: string
splits:
- name: train
dataset_size: 880000000000
OpenGenome2
OpenGenome2 is a large-scale genomic dataset compiled by the Arc Institute and collaborators to train Evo 2, a biological foundation model.
It contains 8.8 trillion nucleotides from across the tree of life, including bacteria, archaea, eukaryotes, bacteriophage, organelles, and curated metagenomic assemblies.
Sequences are provided as text strings (A, T, C, G) at single-nucleotide resolution, optimized for training long-context genome language models.
Composition
- Prokaryotes: 113,379 representative genomes (~357B nt, GTDB v220.0)
- Eukaryotes: 15,032 curated reference genomes (~6.98T nt, NCBI)
- Metagenomic assemblies: 854B nt
- Organelle genomes: 2.82B nt
- Eukaryotic augmentation: 602B nt enriched for gene-rich regions
⚠️ Eukaryotic viral genomes are excluded to mitigate dual-use concerns.
Dataset Features
| Field | Type | Description |
|---|---|---|
text |
string | Raw nucleotide sequence (can be millions of bp long) |
Example Usage
from datasets import load_dataset
# Load in streaming mode
ds = load_dataset("arcinstitute/opengenome2", split="train", streaming=True)
# Inspect first example
example = next(iter(ds))
print(len(example["text"]))
print(example["text"][:100]) # first 100 bp
Citation
Please cite the following when using OpenGenome2:
@article {Brixi2025.02.18.638918,
author = {Brixi, Garyk and Durrant, Matthew G and Ku, Jerome and Poli, Michael and Brockman, Greg and Chang, Daniel and Gonzalez, Gabriel A and King, Samuel H and Li, David B and Merchant, Aditi T and Naghipourfar, Mohsen and Nguyen, Eric and Ricci-Tam, Chiara and Romero, David W and Sun, Gwanggyu and Taghibakshi, Ali and Vorontsov, Anton and Yang, Brandon and Deng, Myra and Gorton, Liv and Nguyen, Nam and Wang, Nicholas K and Adams, Etowah and Baccus, Stephen A and Dillmann, Steven and Ermon, Stefano and Guo, Daniel and Ilango, Rajesh and Janik, Ken and Lu, Amy X and Mehta, Reshma and Mofrad, Mohammad R.K. and Ng, Madelena Y and Pannu, Jaspreet and Re, Christopher and Schmok, Jonathan C and St. John, John and Sullivan, Jeremy and Zhu, Kevin and Zynda, Greg and Balsam, Daniel and Collison, Patrick and Costa, Anthony B. and Hernandez-Boussard, Tina and Ho, Eric and Liu, Ming-Yu and McGrath, Tom and Powell, Kimberly and Burke, Dave P. and Goodarzi, Hani and Hsu, Patrick D and Hie, Brian},
title = {Genome modeling and design across all domains of life with Evo 2},
elocation-id = {2025.02.18.638918},
year = {2025},
doi = {10.1101/2025.02.18.638918},
publisher = {Cold Spring Harbor Laboratory},
URL = {https://www.biorxiv.org/content/early/2025/02/21/2025.02.18.638918},
eprint = {https://www.biorxiv.org/content/early/2025/02/21/2025.02.18.638918.full.pdf},
journal = {bioRxiv}
}