- Xaira is an integrated biotechnology company built to deliver on the promise of AI to help transform the drug discovery and development process
- The company released a preprint with details of its robust platform, Fix-Cryopreserve-ScRNAseq (FiCS) Perturb-seq, specifically designed for generating high-quality data at scale; and simultaneously provided the largest public release to date of this type of data, X-Atlas/Orion
- Xaira also reported details of its method to identify dose-dependent genetic effects within this kind of data
SOUTH SAN FRANCISCO, Calif.–(BUSINESS WIRE)–Xaira Therapeutics today announced a significant leap forward in developing AI-driven virtual cell models with the release of “X-Atlas/Orion,” the largest publicly available Perturb-seq atlas. This extensive dataset was made possible by Xaira’s concurrently introduced “Fix-Cryopreserve-ScRNAseq” (FiCS) Perturb-seq platform, a highly scalable technology designed for large-scale data generation. These combined advancements are poised to accelerate the development of biological foundation models and unlock new frontiers in biological discovery and therapeutic development.
The rapid progress of single cell technology has spurred the creation of foundation models aimed at deciphering complex biological processes. While these models hold immense potential for creating AI-driven virtual cells, their advancement has been hampered by the scarcity of large-scale, high-quality perturbation data. Xaira’s FiCS Perturb-seq platform, which leverages the Chromium platform from 10x Genomics, delivers the sensitivity, scalability and reproducibility essential for generating high-quality perturbational data. Xaira’s innovative approach solves logistical challenges associated with profiling large numbers of cells, effectively captures perturbation-induced transcriptomic changes and accurately recapitulates known biological pathways and protein complexes. The X-Atlas/Orion dataset itself comprises 8 million cells, targeting all human protein-coding genes, with deep sequencing of over 16,000 unique molecular identifiers (UMIs) per cell, allowing for the discovery of biological phenotypes from a wide range of genetic perturbations.
“This industrialized platform and the Orion dataset will empower scientists to build more predictive models of complex biology,” said Ci Chu, vice president of early discovery at Xaira and senior author of the preprint. “We believe this will help us better understand disease biology and discover drug targets.”
One of the most significant advances is Xaira’s method to detect dose-dependent genetic effects, a more refined way of understanding how gene activity changes with the intensity of a given intervention. Traditionally, scientists viewed Perturb-seq gene knockdowns as an “on” or “off” switch. The Xaira scientists showed that the amount of single guide RNA (sgRNA) detected in each cell can be used to measure how strongly a gene is suppressed, offering a much more detailed picture of genetic function. This methodology offers a refined framework to enhance the predictive power and biological insight of future causal models by incorporating perturbation strength as a continuous variable.
“This platform provides the scale and quality needed to model how cells respond across conditions, which is a crucial step toward training the first generation of virtual cell models,” said Bo Wang, SVP and head of biomedical AI for Xaira Therapeutics. “With this foundation, we’re better equipped to uncover disease mechanisms and design smarter therapies.”
This publication caps a landmark year for Xaira since its launch in April 2024. In October 2024, Dr. David Baker, Xaira co-founder, received the Nobel Prize in Chemistry alongside Drs. Demis Hassabis and John Jumper of Google DeepMind for pioneering AI-driven advancements in protein structure prediction and novel protein design. The Xaira team is working to advance these models while developing new methods that can connect the world of biological targets and engineered molecules to the human experience of disease.
Access the Dataset:
X-Atlas/Orion is now publicly available here: https://doi.org/10.25452/figshare.plus.29190726
Read the Preprint:
FiCS Perturb-seq and X-Atlas/Orion publication: https://www.biorxiv.org/content/10.1101/2025.06.11.659105v1
About Xaira Therapeutics
Xaira Therapeutics is an integrated biotechnology company driving advances in artificial intelligence to learn the language of life and transform how we treat disease. The company seeks to rethink the drug discovery and development process from end-to-end by bringing together leading talent across three core areas: machine learning research to better understand biology, expansive data generation to power new models, and robust therapeutic product development to treat disease. Xaira is headquartered in the San Francisco Bay Area.
Contacts
