FILED FROM SANTA CLARA, CALIFORNIA - A profile of the unglamorous backbone of drug discovery
Here is a fact about modern medicine that almost nobody thinks about: before a drug reaches a human being, it usually passes through a mouse - and that mouse is often not a wild-type animal but a carefully engineered one, its genome edited to carry a specific disease, a missing gene, or a stretch of humanized DNA. Someone has to design that animal. Someone has to build it, breed it, validate it, and ship it. For a large and growing share of the world's research labs, that someone is Cyagen.
Founded in 2006 by Lance Han and headquartered in Santa Clara with major operations in China, Japan and Europe, Cyagen Biosciences did something deceptively simple. It took one of the slowest, most artisanal steps in biology - creating a genetically modified rodent, historically a multi-year academic slog - and turned it into a service you could order. Strategy design, DNA vector construction, model generation, breeding, delivery: pick the whole pipeline or just the piece you need. The pitch is not romantic. It is logistical. And that is precisely why it works.
"Cyagen offers a one-stop solution - from strategy design and DNA vector construction to model delivery."
The scale is easy to miss because the product is invisible to the public. Cyagen is now one of the world's largest providers of custom-engineered mouse and rat models. Its catalog holds more than 18,000 validated strains spanning 20-plus research areas - oncology, neurology, metabolism, immunology, rare disease. And the citation footprint is genuinely large: Cyagen's models have appeared in over 10,000 published papers. Most companies would kill for that kind of academic distribution. Cyagen got there by being the reliable vendor, not the famous one.
The bottleneck is time, and time is the whole experiment
If you want to understand why researchers pay for this, you have to understand the enemy: generations. Building a knockout mouse the traditional way means breeding across multiple generations, and mice do not breed on a startup's timeline. Cyagen's answer is a proprietary technology called TurboKnockout, which trims two full generations off the process. A conditional knockout, a reporter knockin, or a humanized model that might have taken years can arrive in roughly four to eight months. In preclinical science, that is not a convenience. That is the difference between running an experiment this grant cycle or the next one.
The company has pushed that logic to its natural conclusion with HUGO-GT - the Humanized Genomic Ortholog for Gene Therapy program - which replaces a mouse's own gene, in place and at full genomic length, with the human version. The point is fidelity. If you are testing an ASO, a CRISPR edit, or an siRNA therapy that depends on the exact human sequence, a rough approximation will lie to you. A full-length humanized model is as close as you can get to a human gene without a human.
"TurboKnockout reduces breeding time by two generations - shortening the cycle to as little as four months."
From ordering mice to manufacturing therapies
Then there is VectorBuilder, which deserves its own headline. It began, modestly, as an online tool for designing viral vectors - the molecular delivery trucks that carry genetic payloads into cells. Researchers liked the tool. They kept asking for more. So Cyagen followed them up the value chain, and VectorBuilder grew into a full contract development and manufacturing organization producing tens of thousands of custom AAV and lentiviral vectors for labs and drug developers worldwide. In January 2025, a VectorBuilder partner, Ucello, won FDA clearance to begin human trials of a CAR-T cell therapy, with VectorBuilder providing the end-to-end CDMO backbone. A design tool had become clinical infrastructure.
VectorBuilder also keeps winning things. It collected a BioTech Breakthrough award in late 2024 and two IMAPAC excellence awards - one for cell and gene therapy, one for biologics CDMO work - in 2025. That same year it launched miniVec, a next-generation plasmid system that strips out the antibiotic-resistance genes conventional plasmids carry. It sounds like a footnote. It is actually a stacked win: safer product, higher manufacturing yield, lower cost. The best improvements in this field often look like subtractions.
The quiet rigor
What is easy to overlook about a vendor like this is how much the trust is earned in details customers never see. When a 2025 Nature study warned that using non-littermate controls can generate false positives because of subtle genetic drift between colonies, the correct response was not a press release. It was to prioritize shipping littermate controls by default. Doing the rigorous thing quietly, before anyone asks, is how you keep scientists as repeat buyers. In a business where a bad control can invalidate a year of someone's work, reliability is the entire brand.
Cyagen is not going to trend. It sells mice, cells and vectors to people who publish in journals, not feeds. But the next time you read about a promising therapy heading into trials, remember that somewhere upstream, someone had to build the animal it was tested in - and increasingly, the answer to "who built it" is a 900-person company in Santa Clara that made the extraordinary feel like ordering off a catalog.
