Basecamp Research, a frontier AI lab harnessing evolution to design new medicines, has announced the first AI models capable of programmable gene insertion, offering a new way to replace faulty genes and reprogram cells for therapeutic use.
Trained in collaboration with NVIDIA, these models are driving the development of a new generation of treatments for cancer and inherited diseases.
In parallel, Basecamp Research secured an investment from NVentures (NVIDIA's venture capital arm) in its pre-Series C round following years of close technical collaboration, which will help the company to accelerate its research and development efforts.
"We believe we are at the start of a major expansion of what's possible for patients with cancer and genetic disease," said John Finn, Chief Scientific Officer at Basecamp Research.
"By using AI to design the therapeutic enzyme, we hope to accelerate the development of cures for thousands of untreatable diseases, potentially transforming millions of lives."
Programmable gene insertion
Programmable gene insertion (placing large therapeutic DNA sequences at precise locations in the human genome) has been a central goal in genetic medicine for decades.
Existing CRISPR-based approaches can only make small edits and must damage DNA to do so, limiting where and how they can be used.
Basecamp Research is the first to demonstrate that AI can design enzymes capable of performing large gene insertions at defined sites in the human genome, opening a long-sought path toward programmable therapies.
Basecamp Research's AI-Programmable Gene Insertion (aiPGI) platform is powered by EDEN, a new family of evolutionary AI models developed with NVIDIA, trained on BaseData, the company's proprietary genomics dataset — the largest of its kind.
The models learn the language of DNA and patterns of evolution, allowing the algorithms to design new, programmable therapies for cancer and genetic disease.
In lab results published in a paper co-authored by NVIDIA, Microsoft and leading academics, the EDEN models designed multiple active insertion proteins for 100% of tested disease-relevant target sites in the human genome, requiring only the genomic target site as a prompt.
This marks a significant step forward in AI model capability.
Basecamp Research has already demonstrated insertion at more than 10,000 disease-related locations in the human genome, including therapeutically relevant integration of cancer-fighting DNA into primary human T cells at novel safe-harbour sites.
This produced CAR-T cells that show strong killing of cancer cells, showing more than 90% tumour-cell clearance in laboratory assays.
AI-designed molecules to fight "superbugs"
In another key frontier therapeutic design task targeting the global drug-resistance crisis, the same model proved its versatility by designing a focused library of novel antimicrobial peptides (AMPs) — small proteins with the potential to kill harmful bacteria — with 97% of candidates demonstrating confirmed activity in laboratory tests.
In a collaboration with University of Pennsylvania scientists led by Prof. César de la Fuente, the top-performing AMPs showed high potency against critical-priority, multidrug-resistant pathogens, offering a powerful new tool in the fight against dangerous "superbugs."
Therapeutic assets in development
These capabilities underpin Basecamp Research's emerging pipeline of cell and gene therapies, opening the path to treatments that are more precise, predictable and personalised than those available today.
The company's goal is to develop potentially curative therapies across a range of cancer and genetic disease indications, powered by continued improvements to BaseData, the EDEN models and aiPGI.