X-Atlas/Orion (Xaira Genome-wide Perturb-seq)

Largest public genome-wide Perturb-seq dataset at release (June 2025). 8 million cells with >16,000 UMIs/cell (~10x deeper sequencing than other atlases). Features dose-dependent genetic effect detection via Xaira's FiCS platform. Designed for training biological foundation models.

Composite
91.4
Experimental validation
Wet-lab confirmed
Stages
Target ID
Modalities
genetic_perturbationsingle_cell
Task types
perturbation_predictiongene_functionfoundation_model_training
Size
cells: 8,000,000
description: genome-wide perturbations
License
CC-BY-4.0
First release
2025-06
Last updated
2025-06
Official site
→ project page
Leaderboard
→ leaderboard
Dataset
→ dataset
Code / GitHub
→ repository
HuggingFace
→ HF
Paper
X-Atlas/Orion: Genome-wide Perturb-seq Datasets via a Scalable Fix-Cryopreserve Platform for Training Dose-Dependent Biological Foundation Models · · 2025 · paper · doi:10.1101/2025.06.11.659105 · 25 citations
Flags
foundation_model_dataindustry_generated
Experts
Groups
Hosted by
Related benchmarks
scPerturb, Open Problems: Perturbation Prediction, X-Atlas/Pisces (25.6M Cell Multi-Context Perturb-seq)

Rubric (7-criterion)

rigor
5
coverage
4
maintenance
4
adoption
4
quality
5
accessibility
5
industry_relevance
5

Notes

Generated by Xaira Therapeutics. Unprecedented sequencing depth enables detection of subtle perturbation effects. Superseded in scale by X-Atlas/Pisces (25.6M cells, March 2026).

← Back to all benchmarks

Compare:
Open comparison →