Biggest Gene Data Drop Ever! 🧬

Dear reader,

If we want to merge AI and biotech, we need data. And freshly last week, the biggest gene data has dropped!

Let’s get into it →

Xaira Drops Massive Perturb‑seq Dataset: X‑Atlas/Orion

AI-driven biotech firm Xaira Therapeutics has released X‑Atlas/Orion, the largest publicly available genome-wide Perturb‑seq dataset. Built using its scalable FiCS platform, the resource includes 8 million single cells across all human protein‑coding genes, with deep sequencing (~16,000 UMIs per cell) and dose-dependent perturbation measurements—marking an unprecedented scale for genetic effect profiling

My take: If you haven’t followed Bo Wang, you need to follow Bo Wang. He was the guys I have been following and he did a great job releasing scGPT (single-cell GPT) and more works on this. Xaira is just a tip of the iceberg. This is a game-changer—by democratizing vast, high-resolution data and embracing dose-response dynamics, X‑Atlas/Orion empowers both AI researchers and wet‑lab scientists to simulate and prioritize experiments with unparalleled precision.

Takeaways

Why It Matters

Scale & open access: Vast data enabling the development of “virtual cell” AI models

Nuanced insight: Moves beyond binary gene edits by capturing continuous, dose-dependent changes

Boosts drug discovery: Models built on this dataset can simulate gene behavior under varied conditions—accelerating target identification and reducing lab workload

Read more: Source