How is it possible to encode things like video and text on DNA, and why would researchers attempt such a feat?

Earlier this year, Swiss researchers reported they developed a technique for storing text, audio, images and video for millions of years, coded into DNA and embedded in glass spheres.

Bioinformatics expert Associate Professor Jonathan Keith of Monash University says such efforts are being driven by the fact that current methods of storing data have a finite life.

Paper and microfilm might survive for over 500 years, but information on CDs and computer disks can often be corrupted, especially as they need to be updated into different formats as technology changes.

"Electronic media is not necessarily safer than paper media," says Keith.

So as human civilisation generates ever more quantities of data, scientists are working on ever more reliable means of storing it long term.

Since the 1990s, a handful of papers have reported the effort to preserve our archives by coding it into DNA molecules.

"All digital images, videos, audio files and text are reduced to strings of zeros and ones," says Keith. "With DNA you have four different bases that make up the molecule so you have a four-character code instead of a two-character code. But the principle is the same."

A computer is used to translate what is required — whether it be a colour, a position, or a letter in the alphabet -- into a particular sequence, using the four-character DNA code.

No living organisms are involved in creating the DNA code. Rather, synthetic DNA molecules with the sequence of required bases are created from scratch.

After a period of storage, a computer is then used to decode the data.

Why store information on DNA?

Storing information on DNA might seem a bit leftfield but, as Keith says, DNA has been used as an "information storage device" in living organisms for millions of years.

And ancient DNA of woolly mammoths, bears and humans dug out of the permafrost suggests DNA has the ability to last a very long time in cold storage — tens to hundreds of thousands of years.

"If you think of it in that way it then it becomes natural to try and use that molecule for our purposes as well," says Keith.

Also, because DNA is a molecule, it takes up far less space than other storage formats.

"The human genome is three billion bases in length and that is stored as a small number of molecules in every cell nucleus," says Keith.

"We're talking about vast amounts of data crammed into minute volumes."

One cup of DNA could store 100 million hours of HD video, say Nick Goldman and Ewan Birney of the European Bioinformatics Institute of the European Molecular Biology Laboratory.

In 2013, their team reported the successful storage of 739 kilobytes of data in DNA — including a colour image, Shakespeare's 154 sonnets, an excerpt from Martin Luther King's "I have a dream" speech and the classic 1953 paper on DNA structure by Watson and Crick.

How long will information on DNA last?

Goldman and Birney suggest the technology they developed could be eventually be used to store data for up to 50 years.

In February 2015, another team led by Dr Robert Grass from the Institute for Chemical and Bioengineering at ETH Zurich encoded the Swiss Federal Charter from 1291 and the English translation of the ancient Archimedes Palimpsest on "The Methods of Mechanical Theorems" into DNA.

While this is only 83 kilobytes of data, Grass and colleagues say they found a way to store the DNA for millions of years.

Keith says the Swiss team's paper provides two advances. First, they have stored the information with 'error correcting codes'.

"It includes some redundancy in the coding so information can be recovered even if some corruption of the stored information occurs," he says.

"This extends the lifetime of data storage because it means we can keep it once the DNA starts to degrade."

Second, Grass and colleagues have stored the DNA-encoded information in 'synthetic silica fossilisation technology' - in other words glass spheres.

"The data is very stable when it is stored in that form," says Keith.

To test the reliability of this encoding and storage method Grass and colleagues heated the DNA-encoded data encased in glass spheres to 70°C for one week and found they could still recover the original data error free.

"The rate of which the data degrades depends on the temperature," says Keith.

Their study suggests data could be stored for 2000 years at 9.4°C or for two million years at - 18°C in the Global Seed Bank in Svalbard, Norway.

"Both papers are significant advances," says Keith. "Neither of them gives us technology we're going to go out and buy at the supermarket tomorrow but both of them are big steps towards a functional very long term storage device."

When will DNA storage be widely used?

Before DNA data storage can be more widely used researchers need to work out how store multiple megabytes of data — the aim is to store zettabytes (sextillion or 10 to the power 21 bytes), says Keith.

And the cost of encoding and synthesising DNA needs to come down — in 2013 it cost US$12,400 for each megabyte of data.

But, Keith expects in a decade costs significant advances could be made on both fronts.

"These things move exponentially rapidly."

For now, storing data in the form of DNA will not be for the everyday person.

"At the moment it looks like the technology is going to mainly be useful for important data that needs to be stored for a long time but don't have to be accessed frequently."

This may include important government and cultural information, but Keith believes a major application will be storage of data generated by scientific projects.

He cites the Large Hadron Collider, which generates a staggering 15 petabytes (1000 terrabytes or 10 to the power 15 bytes) of data per year on its own.

"It generates huge amounts of data but a very limited number of researchers actually work with that, and it's the kind of thing where a decade from now they might come back to data generated recently looking for something in particular."

Associate Professor Jonathan Keith spoke with Anna Salleh

Tags: information-technology, dna

Published 24 August 2015