Cellular Intelligence: Engineering Biology to Design Future Medicine

BOSTON, MA – January 07, 2026 – In a move signaling a profound ambition to reshape medicine, AI-native biotech company Somite today announced it has become Cellular Intelligence. The rebrand coincides with the release of a detailed white paper outlining its mission: to build the first universal virtual model of cell signaling, effectively creating a navigation system for human biology.

This isn't just about mapping the cellular landscape, a goal pursued by many in the field. Cellular Intelligence aims to understand and direct the traffic. By decoding the intricate language of signals that cells use to communicate, the company hopes to transform biology from a science of observation and chance into a predictable engineering discipline.

"Cells respond to signals in intricate ways that govern both health and disease, but we still do not understand the rules that determine these outcomes," said Micha Breakstone, Ph.D., Co-Founder and CEO of Cellular Intelligence. "While the industry races to map the static geography of cells, we are building the navigation system to guide them. That is the shift our new name represents. We are building a future where biology is no longer destiny, but design."

The Engine of Discovery

At the heart of this audacious goal is a proprietary technology platform that gives the company an unprecedented advantage in data generation. Scientific Co-Founder Olivier Pourquié, Ph.D., a distinguished professor on leave from Harvard Medical School, described the core innovation.

"Cellular Intelligence has built a proprietary capsule platform that enables cell-signaling analysis and data generation at scale, with 1,000x higher efficiency than existing methods," Pourquié stated. This efficiency allows the team to run massive parallel experiments, testing millions of unique signal combinations on pluripotent stem cells as they differentiate.

The scale is staggering. According to the company's new white paper, Engineering Cell Fate: Towards a Foundation Model for Virtual Cell Signaling, an early pilot screen of 27,000 unique perturbation sequences was successfully scaled to over 1 million sequential signal combinations in 2025. This firehose of high-fidelity, context-rich data is the fuel for the company's AI models.

Arjun Raj, Ph.D., a tenured professor on leave from the University of Pennsylvania to serve as Head of Computational Biology, emphasized the significance of this milestone. "We have successfully executed some of the largest combinatorial signaling experiments the field has seen," he explained. "This leaves us at a critical inflection point: our data engine is operating at the scale required to unlock the power of machine learning techniques, with massive, high-fidelity datasets yielding more powerful and generalizable models than ever before."

This data deluge is not just about quantity. The company reports that as data density increases, so does the resolution of its biological insights. Where early experiments could only identify the emergence of generic "muscle precursors," the platform can now distinguish between separate clusters for skeletal versus cardiac muscle lineages. This ability to capture the temporal dynamics of cellular change—seeing the journey, not just snapshots of the destination—is crucial for understanding the vast array of intermediate states that define health and disease.

From Lab Bench to Engineering Discipline

The ultimate vision is to move beyond trial-and-error discovery. By understanding the fundamental rules of how combinations, doses, and the timing of signals drive cell behavior, Cellular Intelligence aims to rationally design therapeutic outcomes. Both healthy and diseased cells use the same toolkit of approximately 20 core molecular signaling pathways; the difference lies in how that toolkit is used. By mastering this grammar, the company believes it can write new biological stories.

The potential applications are vast. In regenerative medicine, the platform could enable the precise and efficient creation of specific cell types needed to repair damaged tissues, a long-sought goal for treating conditions from Parkinson's disease to diabetes. In pharmacology, the virtual model could predict how a specific patient's cells might respond to a new drug, flagging potential toxicity or efficacy issues long before clinical trials. It could also computationally model disease mechanisms, offering new targets for intervention.

To achieve this, the company employs a sophisticated active learning strategy. Instead of randomly exploring the near-infinite space of signal combinations, its AI model identifies gaps in its own understanding and designs new experiments to fill them. This creates a self-improving feedback loop, making the discovery process exponentially more efficient over time.

Navigating a Crowded and Ambitious Field

Cellular Intelligence is not alone in its pursuit of a "virtual cell." The concept has become a holy grail in biotechnology, attracting immense talent and capital. Competitor Xaira Therapeutics, for example, is tackling the problem from a gene-centric perspective. Broader calls from organizations like the Chan Zuckerberg Initiative (CZI)—itself an investor in Cellular Intelligence—along with Stanford and Genentech, advocate for a global effort to build a similar model.

What sets the Boston-based firm apart is its laser focus on signaling—the dynamic control system of the cell—rather than the static genetic blueprint. This approach is backed by a powerhouse team and over $62 million in funding from a syndicate of investors that speaks volumes about the perceived potential. Khosla Ventures is known for backing disruptive, category-defining technologies. CZI's involvement underscores the project's deep scientific merit and potential for societal impact. The inclusion of AMD Ventures hints at the immense computational power required and the strategic importance of building the hardware infrastructure to support this new era of biology.

This strategic positioning, combined with the leadership of repeat AI entrepreneur Breakstone (whose previous company, Chorus.ai, was acquired for $575M) and world-renowned scientists like Pourquié and Raj, gives Cellular Intelligence a formidable foundation as it seeks to chart this new territory.

The Promise and Peril of Designing Life

The mission to transform biology into an engineering discipline carries profound implications that extend beyond the laboratory. The ability to "predict and control cell behavior" opens a door to a future of medicine previously confined to science fiction. It offers hope for intractable diseases by providing a rational framework for designing cells that can repair, regenerate, and restore function.

However, this power also brings a new set of ethical and societal questions to the forefront. Phrases like "designing biology" inevitably raise concerns about unintended consequences and the boundaries between therapy and enhancement. As these technologies mature, society will face complex debates about equity and access, ensuring that such revolutionary treatments do not become the sole province of the wealthy.

Regulatory bodies like the FDA will also face the challenge of evaluating therapies born from a process that is dynamic, computational, and predictive, rather than static and descriptive. For now, Cellular Intelligence is focused on the foundational science, building the engine that could one day power these future breakthroughs. The company's work represents a pivotal convergence of artificial intelligence and cellular biology, a turning point that may fundamentally redefine our relationship with the building blocks of life itself.