Seattle AI BioDesign Initiative Targets $95M Protein Engineering
Allen Institute, UW, and Fred Hutch will use machine learning to create biological molecules that evolution never produced.
Three Seattle research institutions have launched a five-year, $95 million initiative to engineer biological molecules that don't exist in nature, using artificial intelligence to accelerate what evolution accomplishes over millennia.
The AI BioDesign project, announced Thursday, brings together the Allen Institute, the University of Washington, and Fred Hutch Cancer Center in an effort to design custom proteins and genetic tools with applications ranging from cancer therapeutics to plastic-degrading enzymes.
Why it matters
The collaboration represents a significant bet that AI can move beyond analyzing existing biology to actively designing new molecular machinery. Success could compress decades of trial-and-error protein engineering into iterative cycles measured in months, potentially unlocking treatments and materials that natural selection never explored.
How the funding breaks down
The Fund for Science and Technology, a nonprofit foundation established by late Microsoft co-founder Paul Allen's estate, is backing the project with $46.1 million going to the Allen Institute, $43.8 million to the University of Washington, and $4.7 million to Fred Hutch Cancer Center.
The AI-biology feedback loop
The project operates on a continuous improvement model: AI systems propose novel biological designs, laboratory scientists synthesize and test those molecules, then feed experimental results back into the models. Each iteration should produce more accurate predictions and more functional designs.
This approach aims to explore biological possibilities outside evolution's path while revealing fundamental principles about how molecular structures determine function.
David Baker, a UW researcher who won the 2024 Nobel Prize in Chemistry for computational protein design, noted that while evolution generates biological diversity, "it is a slow process." The AI BioDesign project seeks to accelerate that timeline dramatically.
Initial research targets
According to Jay Shendure, the project's lead scientific director, early priorities include custom proteins engineered to bind disease markers, genetic switches capable of activating or silencing specific genes, and molecular tools that can selectively degrade or stabilize target proteins.
These capabilities could enable precision medicine approaches, allowing doctors to target disease mechanisms with molecule-level specificity.
From computational models to real-world applications
The project's ambition extends beyond academic discovery. Researchers aim to translate AI-generated biological designs into practical medicines, industrial materials, and environmental technologies that function outside the laboratory.
The details were first reported by Axios Seattle.
This is an original analysis by the Omega editorial team. Source reporting: AI Watch.
Want systems like this working for your business?
Book a Call
