Industrial biology · DNA data storage
Kern Systems
A Boston company spun out of George Church’s Harvard lab to write digital data into DNA with enzymes instead of chemistry. The founding science is peer reviewed. The company built on it has stayed tiny, and its public record goes quiet after early 2023.
A hard drive made from molecules
The insight that started Kern Systems is that a storage medium does not have to be spelled correctly.
Commercial DNA synthesizers chase single base accuracy, because biology punishes a typo. Phosphoramidite chemistry was built for that job. It is slow, it caps out near a few hundred bases, and it produces toxic waste.
Data does not care. A file survives a strand full of mistakes as long as the errors are bounded and the decoder knows it. So Henry H. Lee, then a postdoc in Church’s lab at Harvard Medical School, dropped the accuracy requirement and used a naturally sloppy enzyme instead.
The enzyme is terminal deoxynucleotidyl transferase, or TdT, a template independent polymerase that immune cells use to scramble antibody genes. It grabs the end of a chain and adds whatever nucleotide is nearby. Left alone it adds dozens. Lee paired it with apyrase, which eats the free nucleotides, so the two compete and each cycle adds a short run instead of a long one.
Information then sits in the changes between letters, not the letters. A strand reading AAAATTTGGGGCC collapses to ATGC and carries three transitions. In 2019 the team wrote 144 bits this way, addressing included, and read it back with streaming nanopore sequencing. A codec recovered files from strands each carrying roughly thirty percent error, and a kilobyte scale design stored one bit per nucleotide.
That paper landed on June 3, 2019. Weeks later the company was in Y Combinator’s Summer 2019 batch. Its site still sums up the product in six words: a hard drive made from molecules.
Why anyone wants this
The case for DNA is not speed. It is density and time. The Wyss Institute puts DNA at least a thousand times denser than the most compact solid state drive and at least three hundred times more durable than the most stable tape. Roughly eighty percent of the world’s data is rarely touched after it is written, which is exactly the tier DNA suits.
The case against it is the price of writing. In 2016 the Wyss put chemical synthesis at $3,500 per megabyte. A decade later a Blocks and Files analysis by Philipp Antkowiak, who wrote his doctorate on the subject, put it near $100 per megabyte, still more than a million times what an SSD charges and more than 2.5 million times what tape charges.
Governments noticed before investors did. In September 2019 IARPA launched its Molecular Information Storage program, aiming at a tabletop machine that writes a terabyte and reads ten terabytes a day for under $1,000 and under a kilowatt. Two consortia won contracts worth roughly $48 million: one led by the Broad Institute with DNA Script and Harvard, one led by the Georgia Tech Research Institute with Twist Bioscience and the University of Washington. Kern was in neither.
A lab, a founder, and a handoff
Church is the co-founder whose name opens doors. He encoded his own book into DNA in 2012 and has run DNA data storage as a technology area inside the Wyss ever since.
The working founder was Lee. Harvard’s Church lab directory lists him as a postdoc from January 2012 to December 2020 in the genetics department, project recorded as Vibrio natriegens and DNA information storage. Harvard’s Office of Technology Development, describing the licensed technology, said the lab’s enzymatic approaches cut synthesis cost by eight orders of magnitude and named Kern as the startup emerging from the lab to commercialize them.
In May 2020 Genetic Engineering and Biotechnology News introduced Lee as co-founder and chief executive and let him lay out the method in his own words. That is the last substantial independent interview anyone published with the company. Today the site lists Pradeep Bugga as chief executive, with Lee as co-founder and advisor. No announcement of the change was found.
Then it got quiet
The company news feed has three entries. June 2019: the foundational paper. December 2021: Kern joins the Pagliuca Harvard Life Lab, the university wet lab incubator in Allston. February 2023: the DNA Data Storage Alliance. Nothing since.
The field did not stand still. In October 2020 Church’s group drove TdT with ultraviolet light through a photolithographic mask, building twelve strands in parallel on 1.2 square millimeters and encoding two measures of the Super Mario Brothers theme. That work was funded by the Wyss and authored by the lab, not by Kern. In May 2025 Twist Bioscience spun its storage division out as Atlas Data Storage with $155 million in seed funding.
What is proven, and what is still claimed
| Evidence | What the record shows | Source type |
|---|---|---|
| Founding paper | Lee, Kalhor, Goela, Bolot and Church, Nature Communications 10:2383, June 3, 2019. TdT held in check by apyrase; 144 bits including addressing written and read back by nanopore; a codec validated on strands with about thirty percent error; a kilobyte scale design at one bit per nucleotide. Preprint on bioRxiv, June 2018. | Peer reviewed |
| Follow on paper | Photon directed multiplexed enzymatic DNA synthesis, Nature Communications, October 2020. Twelve strands on a 1.2 square millimeter array. Funded and authored through the Wyss and Church lab. Kern is not credited. | Peer reviewed |
| Federal awards | No contract, grant or SBIR to Kern in USAspending for 2007 to 2026, none in NIH RePORTER, none in NSF. MIST money went to the Broad led consortium (up to $23 million) and the GTRI led SMASH team (up to $25 million). | Not found |
| SEC filings | No registrant named Kern Systems in EDGAR company search and zero Form D hits in full text search as of September 23, 2026. No investment vehicle named for the company. | Not found |
| Patents | PitchBook lists EP-4791538-A1 and WO-2025080581-A1, both titled Parallelized electrochemical polynucleotide synthesis, first filed October 10, 2023. PubChem confirms the publication exists. The applicant could not be confirmed in an open database, and no granted US patent naming Kern was found. | Data provider |
| Industry alliance | The site announces that Kern joined the DNA Data Storage Alliance on February 1, 2023. Kern does not appear on the alliance current members page as of September 2026. | Differs from roster |
| Scale and customers | Kilobyte scale is the largest published demonstration. The 2020 goal of megabytes and beyond has produced no published result. No customer or revenue named anywhere. PitchBook lists two employees. | Not found |
Read plainly: the science is real and published, and it belongs to a Harvard lab that Kern has a claim on rather than to Kern alone. What never appeared is the second thing a storage company needs, a machine. No federal contract, no securities filing, no confirmed patent grant, no customer, no result past a kilobyte. That fits quiet research seven years in. It also fits a company that has stopped. The record does not separate the two.
What to watch
- Whether the October 2023 patent family publishes with Kern Systems named as applicant.
- A first Form D, which would be due within fifteen days of a first sale in any Regulation D round.
- Any demonstration past kilobyte scale, the threshold the founder himself named in 2020.
- Whether write costs across the category fall far enough to make DNA storage investable at all, which matters more to Kern than anything Kern does alone.
In their words
“DNA-based storage does not require writing DNA with single-base precision or accuracy, so we can use rapid enzymatic synthesis that generates nonperfect sequences.”
Henry Lee, co-founder and CEO, Genetic Engineering and Biotechnology News, 2020 · Independent
“We decided to store information in transitions between nonidentical bases to reduce the need for perfect synthesis.”
Henry Lee, Genetic Engineering and Biotechnology News, 2020 · Independent
“Our ultimate goal is to synthesize at incredible scale with tunable accuracy.”
Henry Lee, Genetic Engineering and Biotechnology News, 2020 · Independent
“This study and other advances in enzymatic DNA synthesis will push the envelope of DNA writing much further and faster than chemical approaches.”
George Church, co-founder, Wyss Institute, 2020 · Institution release
“we’re only producing enough new tape to archive 0.3 percent of data produced annually, and more than 99 percent of new data couldn’t be retained even if we wanted to”
David Markowitz, IARPA MIST program manager, Popular Science, 2022 · Independent
“the question is no longer whether DNA storage works; it does. The real challenge is how to make it affordable”
Philipp Antkowiak, DNA storage researcher, Blocks and Files, 2026 · Independent
Related companies
Sources
- Public recordTerminator-free template-independent enzymatic DNA synthesis for digital information storage
- Public recordMolecular Information Storage program page and launch notice
- Public recordAward search for Kern Systems
- Public recordCompany and full text search for Kern Systems
- Public recordOrganization search for Kern Systems
- IndependentDNA Digital Data Whets Its Appetite, by Kathy Liszewski
- IndependentUsing DNA for data storage, by Steve Nadis
- IndependentCan these government efforts crack the code for DNA storage?, by Sarah Scoles
- IndependentDNA Data Storage: When the physics work but the economics don’t
- IndependentTwist Bioscience launches Atlas to bring DNA storage to market
- Data providerKern Systems profile: deals, investors, patents, headcount
- Data providerKern Systems profile: funding, status, locations
- IndependentCurrent members roster
- CompanyKern Systems website: team, news, contact
- InstitutionDigital DNA Storage System technology listing
- InstitutionDNA Data Storage technology page and Enzymatic DNA synthesis sees the light
- InstitutionHenry Lee, lab member listing
Profile researched and written by Healthcare Discovery. Last updated September 29, 2026.
