Gildas Illien

Darling, we need to talk...

Presentation

Paul Bertone

Digital information storage in DNA

The amount of information that humans produce and want to store is increasing exponentially. It is estimated that the total digital information on Earth is of the order of zettabytes (thousands of billions of billions of bytes). The amount of digital information that people want to archive, i.e. store safely, recoverably, for long periods of time with only rare access and with minimal ongoing maintenance requirements, is also growing. However, at present essentially no long-term archiving of digital information is taking place. This is because all current digital storage media require a continual cycle of maintenance to renew both the storage medium and the 'reading' and 'writing' hardware. This in turn is because there is no conventional computing storage technology that is trusted to survive more than a few years. 

Recent genome science-inspired advances in the technologies for reading and writing DNA led us to investigate possibility of using DNA as a digital archive medium. DNA is a stable information carrier, with 10,000-year-old intact sequences routinely recovered from historical samples. Safe DNA storage conditions are easily maintained at low cost, and the ability to read DNA fragments will surely survive for as long as there are technologically-advanced humans inquisitive about the working of living systems. In our proof-of-concept experiment, we showed how existing DNA technologies can be used to store and recover digital information in a manner that could be extrapolated to global data scales, incorporating modern methods such as error correcting codes for data integrity. This talk will describe this experiment, and will speculate on the future of DNA as a digital storage medium.

Carlos Morais Pires

Data Infrastructures in Horizon2020: support to data and computing intensive science