In a paper out this week in Science, researchers Yaniv Erlich and Dina Zielinski report successfully using DNA to store and retrieve “a full computer operating system, movie, and other files”.
DNA has the potential to provide large-capacity information storage. However, current methods have only been able to use a fraction of the theoretical maximum. Erlich and Zielinski present a method, DNA Fountain, which approaches the theoretical maximum for information stored per nucleotide. They demonstrated efficient encoding of information – including a full computer operating system – into DNA that could be retrieved at scale after multiple rounds of polymerase chain reaction.
Which operating system? Turns out it’s KolibriOS, the all-assembler, floppy-based x86 operating system originally based on MenuetOS.
It cost USD9000 and a considerable amount of time just to encode and decode that one floppy worth of data.
DNA copying has a very high error rate (~1:100,000 bases) so it requires a huge amount of redundancy to get 100% accurate reproduction of large data sets.
It is hard to imagine any practical use for the technique.
I’d bet you’d have said the same about transistors.
There is no Moore’s Law for DNA replication. It is an inherently slow biochemical process that almost certainly can’t be sped up by orders of magnitude. [Nature can copy DNA at no more than 1000 bases per second after billions of years of evolution.]
Unfortunately the time to replicate a DNA sequence also increases exponentially with length. So you are realistically looking at a process that takes years to replicate a single TB of data.
Tell me – how long would it take for the ENIAC to replicate a single TB of data?
Moore’s Law ‘works’ because the early transistors were built at densities many orders of magnitude below the theoretical limits. That has allowed continuous improvements for decades.
DNA already replicates at close to the theoretical maximum rate. This is determined by the laws of chemical kinetics. It won’t be altered by spending more money on research.
The rate of DNA replication falls dramatically as the length of the sequence increases. This is because the entire double stranded DNA has to ‘unzip’, copy and and ‘rezip’.
In the bacterium E. coli (4.6 million base pairs) the rate is about 1000 bases per second. In humans (3 billion base pairs) it is only 50 bases per second (20x slower). For a 1TB file (~4 trillion base pairs) the rate is probably going to be much less than one base pair per second. In case you are wondering 4 trillion seconds is over 129,000 years -much slower than ENIAC.
Edited 2017-03-05 09:47 UTC
Maybe, but the research can help in other areas, such has finding a better replacement for DNA. Every little bit helps, and improvements don’t have to be a short-sighted linear extrapolation of what we’re currently doing.
Multithreaded parallel writes buffered in a faster cache?
I really would like you to be wrong on this case but, sadly, you are probably right, as most of us that tried on chemistry practical classes to grow simple organic chains would attest. It is incredibly hard to bond long molecules appropriately and the few techniques I saw (a long time ago) were basically around finding ways to grow the short ones we had. Multi-threaded methods we apply on many other fields to build separated parts and later combine them will, probably, have a hard time on this one.
any new technology was initially very expensive. think back to first hard disks, for instance.
we have to give it a few years and see if it gets any better or not.
plus, dna as very long term storage might just work, even if it is not very quick. data security on the other hand – not necessarily.
Edited 2017-03-06 11:15 UTC
Nothing ever gets better through iteration. Ever.
So basically… In the future, we can theoretically store data in a fingernail. The term “Thumb-Drive” takes a whole new meaning now I guess.
(In the future people will ask, how much your capacity are)
Can’t believe how every other article I’ve read on this has consistently failed to mention that it wasn’t a mainstream multi-gigabyte OS. It puts the whole experiment in context. So yes, you can theoretically store huge amounts of data in a tiny space with a density higher than ever achieved before but they have only actually stored a few kbs worth of data.