Somite AI Secures $47M Series A to Transform Cell Therapy with Foundation Models
核心洞察
Somite AI (搜索) has raised $47 million in Series A funding led by Khosla Ventures (搜索) to develop DeltaStem (搜索), an AI platform designed to revolutionize human cell production for therapeutic applications.
The company's proprietary capsule technology generates cell state transition data at 1000x greater efficiency than existing methods, enabling the development of AI models that can predict optimal conditions for creating specific cell types.
Somite AI (搜索) is targeting multiple therapeutic areas including Type 1 diabetes (搜索), orthopedic conditions, muscular diseases, and blood disorders by developing protocols for producing various human cell types at scale.
Boston-based Somite AI (搜索) has raised over $47 million in Series A funding to advance its AI-driven platform for cell replacement therapy, bringing its total funding to approximately $60 million. The round was led by Khosla Ventures (搜索) with participation from several investors including the Chan Zuckerberg Initiative (搜索), SciFi VC, and Fusion Fund.
The funding will accelerate the development of DeltaStem (搜索), Somite's foundation model platform designed to revolutionize the production of human cells for therapeutic purposes. The company aims to address a broad spectrum of diseases, including Type 1 diabetes (搜索), orthopedic conditions, muscular diseases, and blood disorders.
Reinventing Cell Therapy Development
Somite AI (搜索) was founded just over a year ago with an ambitious mission: to reverse engineer stem cell biology using artificial intelligence. The company's approach could fundamentally transform regenerative medicine by enabling the on-demand creation of any human cell type.
"We're building the foundation model for the human cell," said Dr. Micha Breakstone, Founder and CEO of Somite AI (搜索). "By generating the world's largest cell signaling dataset at 1000x the efficiency of current methods, we're training DeltaStem (搜索) to deliver protocols with unmatched purity, scalability, and reliability."
The challenge Somite addresses is substantial. According to Breakstone, researchers currently know how to reliably produce only about 10 types of cells, while the human cell atlas contains over 5,000 cell types. Traditional methods for developing specific cell types are prohibitively expensive, slow, and rely heavily on trial and error.
Breakthrough Technology for Data Generation
At the core of Somite's approach is a novel method for generating vast quantities of cell development data. The company uses semi-permeable capsules, each containing a few cells with microscopic "windows" that allow biological signals to affect the cells. These capsules are exposed to millions of different conditions, with each cell-condition interaction recorded via a unique barcode.
"What would happen if instead of taking the signals to the cells, we take the cells to the signals?" This question from Professor Allon Klein, a Somite co-founder, led to the development of the capsule technology that generates cell state transition data at unprecedented scale.
Currently, Somite is conducting experiments at approximately a one-million-condition scale and plans to reach 10 million conditions by year-end. This data serves as the foundation for developing AI models that can predict the optimal conditions for creating specific cell types.
The DeltaStem Platform
Somite's DeltaStem (搜索) platform aims to overcome key challenges in current stem cell development: purity (percentage of desired cells produced), scalability (efficient manufacturing at therapeutic scale), and reproducibility (consistent results across batches).
"What AlphaFold (搜索) did by predicting protein structures, transforming structural biology, DeltaStem (搜索) aims to achieve for stem cell biology—predicting the conditions needed to precisely generate mature human cell types at scale," explained Breakstone.
Legendary investor Vinod Khosla highlighted the potential impact: "Somite AI (搜索)'s foundation models, once fully developed and validated, will not only create value for their own pipeline but have the potential to reshape the entire field of human cell therapy."
Therapeutic Focus Areas
With the new funding, Somite AI (搜索) will accelerate key therapeutic programs, including:
- Beta cells for Type 1 diabetes
- Articular cartilage for orthopedic applications
- Satellite cells for muscular diseases
- Hematopoietic cells for blood disorders
The company's initial focus included Duchenne muscular dystrophy (搜索), the most common hereditary neuromuscular disease, but has now expanded to address a broader range of conditions.
A Distinguished Founding Team
Somite AI (搜索) was co-founded by Dr. Micha Breakstone, a seasoned AI entrepreneur who previously founded Chorus.ai (acquired for $575 million), and Dr. Jonathan Rosenfeld, Head of the Fundamental AI Group at MIT.
The scientific co-founders include several prominent researchers:
- Prof. Olivier Pourquié, member of the National Academies of Science and Medicine, Brigham and Women's Hospital and Harvard Medical School
- Prof. Allon Klein, James Prize recipient, Harvard Medical School
- Prof. Jay Shendure, member of the National Academy of Sciences, University of Washington
- Prof. Cliff Tabin, Chair of Genetics at Harvard Medical School, member of the National Academy of Sciences
The Rise of TechBio
Somite represents a growing trend of "TechBio" companies that focus on developing broad technology platforms addressing multiple medical indications, rather than pursuing individual drug development like traditional biotech firms.
These companies take a data-driven approach, developing proprietary AI that reduces costs and risks, shortens development time, and accelerates the shift toward more personalized medicine. The success of companies like Moderna and breakthroughs such as AlphaFold (搜索) have attracted increased venture capital to this intersection of technology and biology.
As Breakstone envisions it, Somite's AI foundation model and platform will help build a future "where we can create any human cell on demand, like a supply of cellular spare parts to repair or replace diseased or damaged tissue."