A $399 kit, a software platform, and a deceptively simple pitch: most DNA tests read a sliver of you. Nucleus reads almost all of it - then tells you what it might mean.
Walk through a New York subway station in late 2025 and you might have read the words "Have your best baby" on a poster. The company behind it was Nucleus Genomics - a roughly 41-person outfit headquartered at 584 Broadway that has spent the better part of a year as the most argued-about name in consumer DNA testing.
Nucleus does something that sounds modest and turns out not to be. It mails you a saliva kit, sequences your whole genome at clinical grade, and hands the result back as readable reports: your risk for common diseases, the rare variants you carry, your genetic odds of living past 80. The catch is in the word "whole." Most at-home tests skim around 0.02% of your DNA. Nucleus reads close to all of it - and that gap is the entire business.
For two decades, sequencing an entire human genome went from a multi-billion-dollar moonshot to something you could buy with a credit card. The cost fell off a cliff. The technology that should have followed - taking all that data and making it useful to an ordinary person - mostly didn't.
The familiar consumer DNA companies built their products on genotyping chips, which sample a tiny, pre-chosen set of common variants. Useful for ancestry and party-trick traits. Largely silent on the rare, serious mutations that whole-genome sequencing can catch. The information existed; the field had simply decided consumers couldn't be trusted with the full picture, or wouldn't pay for it.
Kian Sadeghi started Nucleus in 2021, in his early twenties, after leaving the University of Pennsylvania's computational biology program on a Thiel Fellowship. He has spoken about the family experience that pointed him at the problem - a preventable genetic condition that better information might have flagged. His bet was contrarian: that the reason consumers got fragments of their genome wasn't science, it was caution and business model. Remove both, and people would pay to own and understand all of their DNA.
Investors with a taste for contrarian bets agreed. Peter Thiel, Balaji Srinivasan, and Alexis Ohanian's Seven Seven Six put in early money. Founders Fund, Asylum Ventures, Samsung Next, and One Eight Capital joined a $14 million Series A that closed in January 2025, bringing total funding to roughly $32 million.
The core product is almost boring in its description: a 30x whole-genome sequencing kit you spit into and mail back. What makes it interesting is what sits on top. Nucleus combines your full sequence with polygenic risk scores - which add up the small effects of hundreds of common variants - and folds in lifestyle and environmental inputs to estimate risk for complex, common diseases. Because it reads the whole genome, it can also surface rare, novel, and large structural variants that chip-based tests never see.
Clinical-grade 30x sequencing at home for ~$399, reading roughly 5,000x more of your DNA than a typical consumer chip.
Disease risk, carrier status, traits and ancestry, refreshed over time with a ~$39/year membership.
For couples: the combined risk of passing ~900 inherited conditions to a child.
The contested one: analyzing and ranking IVF embryos across disease risk and traits, with partner clinics.
The skinny gray bar is what most "DNA tests" actually read. The 0.02% bar is widened to 2% here just so you can see it exists - which is, in a way, the whole sales pitch.
Whole-genome sequencing itself is uncontroversial and clinically sound. The argument is about interpretation - specifically the leap from "we sequenced everything" to "we can rank embryos by intelligence or height." When Nucleus extended its platform into embryo selection in 2025, professional groups and scientists pushed back hard. Critics in Scientific American and MIT Technology Review called trait-based embryo ranking scientifically shaky and ethically fraught; some labeled it a new eugenics. A competing firm, Genomic Prediction, filed a trade-secrets lawsuit tied to a co-founder who moved to Nucleus.
Nucleus, for its part, frames the work as giving parents information and choice. The disease-risk and carrier-screening products draw far less heat than the embryo tool. The company is doing two things at once: shipping a genuinely useful consumer sequencing product, and stress-testing how far society will let that data be used.
Strip away the subway ads and the lawsuits and the mission is consistent: make whole-genome sequencing affordable and useful enough that reading all of your DNA becomes a normal first step in preventive medicine, not a specialist's privilege. The membership model - pay once for the sequence, a little each year for new insight - is built on the idea that your genome doesn't change, but what science can tell you about it keeps growing.
As sequencing keeps getting cheaper, the question Nucleus forces is no longer "can we read the whole genome?" - we can - but "what should anyone be allowed to do with it?" That question lands on regulators, fertility clinics, and ordinary buyers, not just one New York startup. Nucleus has made itself the place that argument happens.
Go back to that subway poster. A year earlier it would have read as science fiction or marketing bravado. By the end of 2025 it read as a live policy debate with a checkout button attached. Nucleus didn't invent whole-genome sequencing. It did the more uncomfortable thing: it put the full genome in front of consumers and asked them to decide what it's worth - which is exactly the conversation the field had spent twenty years avoiding.