Researchers build a rudimentary DNA-based computer

For the first time, a multidisciplinary team of researchers have demonstrated a proof of principle of all the capabilities needed for a functioning computer—storing, retrieving, processing, erasing, and rewriting data—using nucleic acids. With this primitive computer, the researchers were able to solve simple chess and sudoku problems (Nat. Nanotechnol. 2024, DOI: 10.1038/s41565-024-01771-6).

Albert Keung, a biomolecular engineer at North Carolina State University (NCSU) and the study’s lead investigator, says, “We’re basically taking the same genetic material that’s in you and me and developing ways to store information and compute information using that material.” Considering that the storage density of DNA is many orders of magnitude higher than that of any manufactured device, this work explores the possibility of creating a new form of computer that can fit all the world’s information in a shoebox, he says.

In DNA-based systems, information is stored as the four nucleotide bases that compose nucleic acids, and the computing occurs in the form of chemical reactions in the DNA. But unlike in silicon-based computing—in which the same substrate can be used to read and write upon—the DNA itself is both substrate and information. Reading or computing on it destroys the DNA. There are ways to immobilize DNA on a substrate and read it directly, but this sacrifices high-density throughput.

This study overcomes these impediments with a soft polymer material developed in the lab of Orlin Velev, a chemical engineer at NCSU, in 2019. The material, called soft dendricolloids, has a hierarchical branched structure at the nanoscale level. “Working together with Albert [Keung’s] group, we found that it is very good for being able to take in the DNA, keep it inside and protect it, but still have it available for reading and manipulation,” Velev says.

The material also allowed them to add magnetic components and move the whole bundle around in a microfluidic environment. “Potentially, in the future, [we should be] able to store it someplace and then take it out and read it,” Velev says.

The material’s high surface area helps it absorb a lot of DNA and preserve its natural information density. It does so in a nonpermanent way, so the researchers could work on the data—copy them and perform computations on them—in a nondestructive way.

“We didn’t have to chemically link the DNA to the material—it just naturally absorbed it. That also makes [the process] reversible,” Keung says. The team could also chemically remove the DNA from the polymer and load a different file onto that same material—which is analogous to how a hard drive works: “You can erase data and put new data on it.”

In their accelerated aging studies, the researchers calculated the half-life of the DNA to 2 million years.

The DNA-based computing demonstrated in this study is reminiscent of rudimentary hard drives with read, write, and erase functions, says James Banal, a chemist at the Massachusetts Institute of Technology who was not involved in the project. “It offers a fascinating glimpse into the possibilities of DNA-based information systems,” he says. The new research is an intriguing proof of principle, he adds, but one that requires substantial further development.