DNA Data Storage: Storing the Internet in a Test Tube
Biotechnology proposes storing digital files in DNA molecules: millions of times denser than a hard drive, with durability measured in centuries. That is the race toward molecular storage.
One Gram of DNA to Store the Entire Internet
While data centers around the world drown in hard drives that expire within a few years, biotechnology proposes an unexpected solution: storing digital information in DNA molecules, the same medium nature has been using to preserve information for billions of years. Recent advances in the field, with milestones from companies such as Microsoft Research, Twist Bioscience, and the consortium DNA Data Storage Alliance, bring this technology back to the center of the debate over how humanity will archive its digital memory.
The bet is as simple as it is ambitious: if data are encoded in the four nitrogenous bases of DNA (adenine, thymine, cytosine, and guanine), a single gram of genetic material could theoretically store around 215 petabytes, the equivalent of millions of conventional hard drives, and remain readable for centuries if kept under proper conditions.
How Information Is Written into a Molecule
The process converts the bits of a digital file (ones and zeros) into sequences of bases, which are chemically synthesized as short DNA fragments. To read them back, ordinary genetic sequencing is used, the same technology that today costs only a few euros per genome and has drastically reduced costs over the past decade.
- Writing: the data are translated into bases and synthesized into DNA strands of 100 to 200 nucleotides.
- Storage: the strands are preserved in silica capsules that protect them from moisture and heat.
- Reading: a sequencer reads the bases and software reconstructs the original file bit by bit.
One of the most counterintuitive advantages of the technique is its durability: while a reliable modern hard drive lasts 5 to 10 years and magnetic tape reaches about three decades, DNA fragments recovered from bones tens of thousands of years old remain readable today.
From the Laboratory to Cold Archives
The first commercial uses will not go to hot data that is consulted every second, but to the massive archiving of cold data, which by several estimates represents more than 60% of the information generated by companies and institutions. Historical backups, medical records, scientific archives, or audiovisual heritage are ideal candidates: they are written once and read rarely.
“The density of DNA exceeds any known physical medium by several orders of magnitude, and that completely rewrites the rules of long-term storage,” researchers in the field often summarize.
Among the milestones that have kept the discipline alive are the complete restoration of an operating system from DNA, the encoding of full films and books into micrograms of material, and molecular archive projects funded by U.S. and European research agencies.
The Challenges Still Ahead
Nobody claims DNA will replace flash memory next year. The obstacles remain considerable: synthesis — writing — is still slow and expensive, reading requires chemical reactions lasting minutes, and random access to a specific file within a tube containing billions of molecules needs sophisticated indexing mechanisms that are only just beginning to mature.
Even so, the trend is clear. The cost of DNA synthesis is falling steadily, and the field's stated goal is to reach a cost per terabyte competitive with tape archiving within the next decade. If they succeed, humanity's hard drive will stop being metallic and become organic, wet, and tiny.