Biology Has An Answer For Massive Data Storage – DNA

Back in 2018, Peter Diamandis of the X-Prize described DNA as a replacement for silicon-based technologies used to store data. He wrote, “At its theoretical limit, we could fit 215 million gigabytes of data in a single gram of DNA.” The way this would work would involve converting binary data bits, the zeroes and ones into the base nucleotides that make up the DNA molecule. The nucleotides are identified by the letters A for adenine, T for thymine, C for cytosine and G for guanine. The four comprise the double helix that forms the instruction manual for building all life on Earth.

For binary data storage, a ’00’ can be encoded as A, ’01’ as G, ’10’ as C, and ’11’ as T. A Scientific American article that appeared on May 28, 2021, noted that a gram of DNA “can archive a staggering amount of information.” It can be stored in salt and remain stable for decades at room temperature. In a controlled environment such as a data centre, DNA storage would require little maintenance and would be easily copied.

How much can be stored within the double helix molecule? The Scientific American article notes that all the information we will generate in 2025, estimated to be 33 zettabytes (3.3 followed by 22 zeroes) could be stored in a volume of DNA no larger than a ping-pong ball. The entire U.S. Library of Congress, some 74 million million bytes “could be crammed into a DNA archive the size of a poppy seed” with lots of room left over to do it 5,999 more times.

According to the Wyss Institute at Harvard University, the current cost to synthesize DNA for data storage is US $3,500 per megabyte. Wyss researchers have attempted to write data directly into DNA using a method called “de novo synthesis.” Described as a low-cost approach using DNA for data storage, it deploys an electronic means of encoding data to the four nucleotide letters.

Thought to be state of the art, the work at Wyss appears to be supplanted by “epi-bit,” an even newer methodology that was recently described in an article appearing on October 23, 2024, in the journal Nature.

This research is a joint collaboration involving Peking University, North China Electric Power University, the Max Planck Institute and Arizona State University. It describes epi-bit as a method to allow the writing of arbitrary data on DNA using premade nucleic acids.  “Through self-assembly guided enzymatic methylation, and epigenetic modifications,” data bits are placed into DNA using “molecular movable-type printing.” Using selective methylation, epi-bit “achieved synthesis-free writing of approximately 275,000 bits on an automated platform.”

Data-encoded DNA will lead to DNA drives that will encode error-free data using epi-bit writing kits. Vials of encrypted DNA will replace massive silicon-based data centres. Instead of warehouses, DNA drives will be housed on a shelf or uploaded to the cloud. Since DNA can last millennia without maintenance why will we need backup and archiving that takes up even more space?

Diamandis describes future space missions carrying multiple DNA storage cubes each no more than 5 cubic centimetres containing a billion gigabytes of data. It won’t take much to transport all the knowledge of humanity in a few cubes of DNA. Imagine what this will mean for future space colonies or for a permanent human presence on the Moon and Mars.

Diamandis notes, “Throughout evolution, DNA has unlocked extraordinary possibilities—from humans to bacteria. Soon hosting limitless data in almost zero space, it may one day unlock many more.”