How to store data on DNA?

Data centers worldwide store several trillion gigabytes of information, with 60-80% classified as cold data—archives, old emails, and legal documents virtually never accessed yet continuously maintained at enormous environmental cost.

Inspired by Richard Feynman’s 1950s proposal to learn from biology, researchers now harness DNA molecules—nature’s proven information storage system that every living cell uses to encode genetic instructions.

Computer scientists and biochemists collaborate to bridge digital and biological worlds, converting electrical signals into molecular arrangements while respecting biochemical constraints and error tolerance requirements.

Molecular biologists synthesize artificial DNA with no biological function—purely physical objects designed to hold data. Researchers cut files into shorter strands and create multiple copies for error-checking through cross-referencing.

Researchers developing passive storage solutions target cold data archives—files requiring preservation but rarely accessed, including legal documents, cultural heritage materials, and scientific records maintained by institutions like the European Parliament and French National Library.

Researchers have developed sequencing technologies since the 1970s, primarily for biology. These techniques recently enabled reading million-year-old mammoth DNA and now serve technological purposes for retrieving stored digital files.

Despite researcher optimism, current DNA storage faces severe practical constraints. In 2024, synthesis remains billions of times slower than hard drives, and costs reach $1,000 per megabyte, prohibiting widespread commercial adoption.

Mark Antonini at France’s i3s laboratory leads research on DNA storage through the PIA MOLÉCULARCHIVE program. He chairs the international JPEG DNA working group, collaborating with institutions like the French National Library, European Parliament, and Software Heritage.