UK United Kingdom

DNA data storage: 100 million hours of HD video in every cup

Biological systems have been using DNA as an information storage molecule for billions of years. Vast amounts of data can thus be encoded within microscopic volumes, and we carry the proof of this concept…

You could fill this with coffee … or the equivalent of millions of DVDs. raindog/Flickr

Biological systems have been using DNA as an information storage molecule for billions of years. Vast amounts of data can thus be encoded within microscopic volumes, and we carry the proof of this concept in the cells of our own bodies.

Could this ultimate storage solution meet the ever-growing needs of archivists in this age of digital information?

This dream has come a step closer to reality with the publication of a new technique in this week’s edition of the scientific journal Nature.

Stored in DNA

A team of researchers headed by Nick Goldman and Ewan Birney at the European Bioinformatics Institute of the European Molecular Biology Laboratory (EMBL-EBI) has dramatically demonstrated the potential of the technique to store and transport human-made data.

Their data included some well-chosen iconic elements: Shakespeare’s 154 sonnets, an audio excerpt from Martin Luther King’s “I have a dream” speech, Watson and Crick’s classic paper on the structure of DNA, and a colour photograph of the European Bioinformatics Institute.

These files, in common digital formats found on almost every desktop computer, were encoded byte-by-byte as DNA molecules, shipped from the USA to Germany without specialised packaging, and finally decoded back into their original electronic formats.

Although the study involved less than a megabyte of data in total, this is already orders of magnitude more than has previously been encoded as synthesised DNA.

The authors argue convincingly that the technique could eventually be scaled up to create a storage capacity far beyond all the digital information stored globally today (somewhere in the vicinity of 1 zettabyte or 1015 megabytes).

Perfect for data storage

DNA molecules are natural vehicles for digital information. They consist of four chemicals connected end-to-end like characters of an alphabet to form long strings similar to a line of text. DNA molecules are even more similar to the sequences of zeroes and ones that digital computers use to represent information.

DNA has substantial advantages over both printed text and electronic media. For one thing, it can remain stable for long periods of time with a minimum of care. Intact DNA has been extracted from bones (and other organic matter) tens of thousands of years old, and its sequence reconstructed with as much detail as if it had come directly from a living organism.

Another advantage of DNA over electronic media is that it requires no power supply to maintain its integrity, which makes it easy to transport and store, and potentially less vulnerable to technological failure.

Dr Nick Goldman of EMBL-EBI, looking at synthesised DNA in a vial. European Molecular Biology Laboratory

Perhaps the greatest advantage of DNA as a storage medium is its minuteness. For example, EMBL-EBI’s official press release claims that more than 100 million hours of high-definition video could be stored in roughly a cup of DNA.

We’re getting there

DNA storage devices won’t be available in the supermarket any time soon. The major drawback is the current cost of synthesising DNA in the quantities required, estimated at around US$12,400 per megabyte of data stored.

This is cost-effective only for archives intended to last hundreds or even thousands of years – something few of us contemplate.

The main cost of maintaining electronic archives over such a long period of time is that the media have to be periodically replaced and the data copied, whereas DNA has merely to be stored somewhere cool, dry and dark.

But if the cost of synthesising DNA can be reduced by one or two orders of magnitude – which, judging by current trends could occur within a decade – DNA archives intended to last less than 50 years would become feasible.

Another issue is the cost of decoding the information stored in DNA, estimated at about US$220 per megabyte. At that price, DNA archives would only be rarely accessed. Yet this too could change in the near future, given the rapid pace of innovation in DNA-related technologies.

We shouldn’t let these practical issues distract from the significance of this exciting innovation.

As the inventors point out, the technique may already be economically viable and attractive for certain long-term, infrequently accessed archives, including some government and historical records, or science projects that generate massive amounts of data.

Examples of the latter include important large-scale experiments in particle physics, astronomy and medicine.

But perhaps the most exciting aspect of this proof-of-concept study is the impetus it provides to further innovation and the unexplored doors of possibility it opens.

Articles also by This Author

Sign in to Favourite

Join the conversation

12 Comments sorted by

  1. Heath Pardoe


    This is really interesting. Am I blind or is this article missing a link to the abstract for the original article?

    1. Matt de Neef

      Editor at The Conversation

      In reply to Heath Pardoe

      Thanks for that Heath. We'll update the link now.

  2. Guy Cox

    logged in via email

    There are a few ifs and buts here ... Yes, we can more or less reconstruct the DNA of an extinct organism, but this is done by looking at fragments and piecing them together by looking for overlaps. Without multiple copies we won't get far. That is also what was done in this paper - data was stored in multiple overlapping short lengths of DNA. With a mere 730KB of data (less than 1 CD) this may be OK, but by the time we are getting into petabytes the computer power needed to piece together the bits would become a bit challenging.

    Scanning-probe techniques offer the prospect of one atom per bit, with inorganic media, and are a far more likely prospect for the future, even if DNA is currently ahead.

    1. Jonathan Keith

      Senior Lecturer, School of Mathematical Sciences at Monash University

      In reply to Guy Cox

      This technique incorporates indexing information into each short DNA fragment, so the computationally expensive assembly algorithms needed for reconstructing DNA from living organisms aren't necessary. Even so, a technology that uses one atom per bit would certainly be a wonderful thing.

    2. Guy Cox

      logged in via email

      In reply to Jonathan Keith

      The current record is 12 atoms per bit. "Bistability in atomic-scale antiferromagnets"; Sebastian Loth, Susanne Baumann, Christopher P. Lutz, D.M. Eigler, Andreas J. Heinrich; "Science", Bd. 335, S. 196, DOI: 10.1126/science.1214131

      However the amount of data stored to date is rather small - 1 byte. (01010011 - ASCII code for the letter S).

  3. Gerald Officer

    Lab rat

    "100 million hours of HD video " = my cup floweth over...

    I already have mutant data.

  4. Fred Moore


    There is a naive assumption here that humans will breed infinitely on this planet in order to create the amounts of data that would require DNA storage. With stable populations the amount of data required would not need DNA S/R.

    Fact is that we have reached the carrying capacity of this planet already and all the lies trying to hide that fact in the name of economic growth (which is nothing more than misplaced male sexual frustration) are essentially the data that DNA systems will be meant to…

    Read more
  5. Alex Cannara

    logged in via Facebook

    "100 million hours of HD video in every cup " -- perfect for the new multi-media caskets available in te US, so our dead beloveds can have their own home-theatre systems playing into eternity. Playlistscan even be controlled externally, by survivors -- hope they don't have grudges and select uncomfortable content, looping for eternity.

  6. Eric Shawn Bosloor

    Public Relations SuperCheap Storage at Super Cheap Storage

    Great article! Thank you for sharing. I am truly excited by the probability of DNA storage. Imagine, all the data we have in our century stored in DNA! That would mean libraries of literature, history, fiction and so many ideas, philosophies and music all in DNA storage, all in the palm of your hand. That would mean almost an infinite amount of hard disk space in your computer too! Just image the games that could potentially be played on that! Oh and then there would be smart cars, smarter phones, and smart aeroplanes because of the vast amount of information in DNA storage. Wow.

  7. Michel Syna Rahme

    logged in via email

    That is pretty cool stuff! Very interesting!