Illumina expands Billion Cell Atlas with Formation Bio, two others

Illumina announced on the date of this article that the Billion Cell Atlas alliance has expanded to include Formation Bio, an AI-native drug developer, plus two additional AI-driven drug discovery partners. The program was introduced in January 2026 with founding members AstraZeneca, Merck, and Eli Lilly and Company.

The Atlas is a genome-wide genetic perturbation dataset designed to capture how cells respond to genetic changes via CRISPR. To date, over 350 million cells have been sequenced, generating upwards of six petabytes of genomic data across hundreds of disease-relevant and healthy cell types. Upon completion, the program will have sequenced one billion individual cells across more than 200 disease-relevant cell lines. The scale and diversity of the dataset can reveal how diseases originate and develop, and potentially how their progression may be reversed.

Rami Mehio, senior vice president and general manager of BioInsight at Illumina, framed the effort as addressing "key bottlenecks across the drug discovery and development continuum." The Atlas enables users to discover and validate novel targets, characterize drug and disease mechanisms of action, and explore potential new indications. For Formation Bio, CEO and co-founder Benjamine Liu explained that the company's model involves identifying and acquiring promising medicines already close to or in the clinic, then using AI to develop them faster and more efficiently. A core part of that model is making better asset-selection and indication decisions, especially for first-in-class programs where evidence is fragmented. By leveraging the Illumina Billion Cell Atlas to credential therapeutic-area hypotheses and target-indication pairs against cell-specific causal biology, combined with genetics, human biology, translational evidence, and clinical data, Formation Bio can improve its ability to choose the right assets, indications, patients, and trial designs—and ultimately increase the probability that important new medicines reach patients.

Kyle Farh, vice president of Artificial Intelligence at Illumina, noted that "the next frontier of AI in biology hinges on the creation of foundational training datasets," adding that up until now most single-cell data has been observational and that Illumina aims to change that paradigm.

The Illumina Billion Cell Atlas represents a shift in how drug discovery leverages biological data at scale. Launched in January 2026, the program was built to address bottlenecks across drug development: identifying what targets to pursue, understanding why drugs succeed or fail, and selecting the right patient populations for clinical trials. The addition of Formation Bio and two other AI-driven companies signals confidence that machine-learning models trained on large, diverse biological datasets can improve the rigor of asset selection—a critical decision point where many first-in-class medicines fail because their target biology is compelling but clinical evidence is sparse.

Formation Bio's stated model—acquiring promising drugs close to or in the clinic and using AI to develop them faster—aligns directly with what the Atlas enables: using single-cell perturbation data to build more precise models of how candidate drugs interact with disease biology and to identify patient subgroups most likely to respond. By integrating the Atlas's cell-state-specific data with genetic, translational, and clinical evidence, developers can more rigorously credential target-indication pairs before investing in expensive clinical trials. This capability is particularly valuable for first-in-class programs, where the clinical precedent is limited and the stakes are high.

No comments yet. Be the first to share your thoughts!