Basecamp Research closed a $140 million Series C to train general-purpose AI models on the Trillion Gene Atlas, a genomic dataset assembled through partnerships across more than 30 countries and all seven continents. Rather than modeling the narrow, heavily studied slice of human and pathogen biology, the company is training on largely unexamined natural sequence diversity to design cell and gene therapies, enzymes and peptides. Its lead application is in vivo cell therapy — reprogramming a patient's own cells inside the body rather than extracting, modifying and reinfusing them.
Key Points
- $140M Series C funds AI models trained on unexplored genomic diversity
- In vivo cell reprogramming targets cost and complexity of current therapies
- Backers include NVIDIA, NATO Innovation Fund, Rockefeller Foundation, Roche vice-chairman
Longevity Analysis
Therapeutic design has been constrained by the narrowness of its reference data — a small, repeatedly mined fraction of known sequence space. Expanding that input layer changes what can be read from biology and, consequently, what can be written back into it, particularly for tools that instruct cells to restore their own repair capacity rather than compensate for its loss. The in vivo approach removes the extraction-and-reinfusion step that makes current cell therapies logistically and financially inaccessible, which determines whether regenerative interventions remain exceptional or become routine.
Original published by Longevity.Technology, by Kyle Umipig.