A Boston-based startup called Catalog has been working for five years to prove that it can use strands of synthetic DNA to store information more efficiently than current computers, and even run computer applications in a test tube.
Now the proofs of concept are complete — including storing the entire text of Wikipedia on DNA — and the company is turning to developing the technology to address real-world problems.
The initial efforts are focused on tasks that require searching through huge sets of data quickly to find desired information. That could mean sussing out fraudulent financial transactions or finding promising spots to drill for oil using seismic surveys, said David Turek, Catalog’s chief technology officer, who joined the company last year from IBM.
“We’ve made discoveries on the bench, if you will, but the issue is how do we translate that into a commercial offering,” Turek said.
Last month, the company raised $35 million in funding from a group of investors led by Hanwha Impact Partners. With the new money, Catalog is doubling the size of its office and lab space in the old Schrafft’s Candy building in Charlestown.
The company was founded in 2016 by MIT scientists Hyunjun Park, the current CEO, and Nathaniel Roquet, who left Catalog last year and is lead scientist at Tessera Therapeutics in Cambridge.
Traditional computers run on chips that are fiendishly complex, squishing billions of circuits onto a piece of silicon the size of a coin. DNA molecules encode genetic data at an even more compressed scale; a single strand of human chromosome includes up to 300 million base pairs.
Catalog’s technology uses the four compounds that make up the base pairs — adenine, cytosine, guanine, and thymine — to encode information onto synthetic strands, much like the electrical impulses that represent blocks of data on silicon chips. A DNA computer wouldn’t need much if any electricity, as chemical reactions could power its computations. And it could store millions of times more information than a hard disk in the same volume of space.
Catalog has shown such breakthroughs are possible but now needs to find ways to make its innovations economically viable and usable on a larger scale. At least initially, the plan is for Catalog to own the DNA storage and computing systems and let customers access them over the Internet, similar to the business model of Amazon’s cloud services.
“You can access our technology over the Web, use it over the Web. ... That way, you don’t have to get your hands dirty,” Turek explained.
Catalog’s efforts come as improvements in computers powered by traditional silicon chips are slowing. That has attracted funding for a variety of new kinds of technologies, from optical chips to quantum computers.
One point to clarify: The DNA strands that Catalog uses are created by machines and don’t carry information about living creatures.
“All the DNA we’re talking about is synthetic, and none of it is or can be made biologically active,” Turek said. “We’re not building DNA for ‘Jurassic Park’ applications, nothing like that.”