When COVID-19 closed down most of society in 2020, Kian Sadeghi, like many of his then-classmates at the University of Pennsylvania, found himself holed up in his bedroom. He prefers it that way.
During those months at home in Brooklyn, Sadeghi filled eight spiral notebooks with research on genetic sequencing, computer science and statistics. The result was the formation of Nucleus Genomics (still listed as a “stealth” startup on his LinkedIn, as of writing), which he has dropped out of Penn to pursue full time.
Nucleus’ Genomics pitch, per Sadeghi, provides a fresher take on genomic analysis. Users can upload their genetic information obtained through an established company, like 23andMe or Ancestry, or order a genetic testing kit. Then the company’s “proprietary algorithms” will calculate polygenic risk scores – measure of disease risk based on analysis of genetic variants – and display them in a user-friendly platform.
“The thesis is a human genome on every iPhone,” Sadeghi told TechCrunch. “So we combine genetic and non-genetic information to eventually calculate the absolute risk scores for people for disease.”
Nucleus announced a $3.5 million seed round on Tuesday led by Founders Fund. The round includes participation from Adrian Aoun (CEO at Forward Health), Patrick Hsu (Professor at UC Berkeley and co-inventor of CRISPR), Brent Saunders (former CEO at Allergan), Matteo Franceschetti (CEO at Eight Sleep), Joshua Browder (CEO, DoNotPay), Cory Levy, Adam Guild, Pareto Holdings, Pomp, Shrug, Ankur Nagpal, Quiet Capital, Salt Fund, Bluewatch Ventures, Night, and Brandon White.
The background behind Nucleus Genomics is the idea that people will be eager to know more about their genomes. And that they’ll want to learn how to actually interpret that information.
Sadeghi clearly has this interest himself. He traces the ideological origin of his company to the premature death of a cousin from long QT syndrome, which in some cases, is genetically linked. He then started to experiment with genes on his own at Brooklyn-based Genspace, a community biology lab. (He was featured in a Wall Street Journal story on DIY CRISPR enthusiasts).
Sadeghi’s interest in genetics is shared by others. Direct-to-consumer genetic testing initially spawned a subculture of people who have uploaded their genetic information to services like GEDMatch, which has been used to help people find lost relatives, or under controversial circumstances, solve cold cases.
But direct-to-consumer genetic testing has truly exploded in recent years. In 2013 fewer than 2 million people ordered consumer genetic tests. By 2018, it was north of 12 million. By 2021, Bloomberg reported that one in five Americans has turned over their genetic data to either 23andMe or a competitor.
There could be more fuel left to burn in this fire. One 2018 poll run by the Associated Press NORC Center for Public Affairs Research suggested as much as 54 percent of Americans had still heard little or nothing about genetic testing.
Sadeghi’s pitch is many of those who have yet to embrace genetic testing are younger people. There’s some evidence to back this up, though not a huge amount. A 2017 survey of about 1,600 DTC genetic test consumers (from 23andMe and Pathway), found that the average user was middle aged, white, and, most likely, a college educated female.
The average age of a 23andMe user, per the company’s 2011 numbers, is 45. TechCrunch has reached out to 23andMe to confirm the age of their user base, but has not heard back.
Nucleus, he says, will aim to boil down genetic testing into user-friendly reports that will be distributed through “newfound distribution channels” – an effort to reach younger audiences.
“We can tap into a brand-new audience and introduce an entire generation to genetic testing, just as you could say Robinhood introduced an entire generation to the stock market,” he said.
Nucleus, though, won’t necessarily rely on new users. It also aims to become a third-party platform, where people could upload their genomic data collected by other companies. There are already third-party platforms like impute.me (a nonprofit), and Allelica (raised 1.75 million in a recent seed round) that provides detailed information on clinical disease risk if users upload genetic information.
Within that landscape, Nucleus does offer a good incentive to upload: cash for the user.
Nucleus will also be open to sharing genetic information with drug companies, says Sadeghi, if users opt-in to sharing their information. Should they choose to share genetic information through Nucleus they can earn a potential “dividend payment.”
Consumer data is already being shared anyway. 23andMe has struck a deal with GSK to use genetic information for drug identification and research – only data from users who opt-in to this service is shared. (23andMe and GSK already have a cancer drug in clinical trials.)
In exchange for the genetic data they provide, which increasingly could be used to develop money-making drugs and other medical products, users get their own payout.
As of now, the amount of that dividend payment is undecided, says Sadeghi.
Regulatory paths
In 2013, 23andMe first debuted reports using genetics to predict health risk. But FDA asked the company to pump the brakes and issued a warning letter. Since then, the company has received FDA approval for a number of genetic tests that quantify health risks for conditions like Parkinson’s or breast cancer, to name a few. A diabetes risk calculation test was debuted in 2019, but didn’t go through FDA approval because it was called a “wellness product”– sparking some criticism over its utility.
23andMe essentially hammered out the process of getting DTC tests and reports approved. And that has opened up the possibility of obtaining 510(k) premarket clearance for other genetic analysis algorithms, should they show equivalent accuracy. (23andMe was concerned about this exact process, and disclosed it as a risk factor in S-1 filings).
Right now, Nucleus is still developing its polygenic risk score algorithms. Those are the company’s secret sauce.
“It’s extremely technical, extremely hard, extremely cutting edge,” Sadeghi said. But as for regulatory strategy, the company initially plans to go a route once used by Ancestry – order tests from a clinical network, rather than creating their own.
That does allow the company to sidestep FDA approval of its algorithms at first (this process is regulated by the Centers for Medicare and Medicaid). But perhaps not for long. Eventually Ancestry pivoted and applied for 510(k) market approval for its own DNA testing kit and genetic test.
When TechCrunch pointed this out in a followup email: Sadeghi said: “Eventually, in the future, we do seek to go that route.”
In the meantime, the company plans to launch sometime around January 2022 and has plans to grow the team from three to six or seven members.
“It’s an imminent inevitability that every person has the human genome sequence,” said Sadeghi. “The question is who is actually going to respond to collect that data? I’m hoping that it’s going to be Nucleus.”