DNA storage
A next-gen storage medium with a biological ingredient: DNA molecules can store digital data with an extremely high density and this for thousands of years.
The DNA double helix structure was discovered back in 1953, by James Watson and Francis Crick. Back then, they couldn’t imagine how it would transform our understanding and diagnosis of diseases, the design of new therapies, and the breeding of resistant and more productive crops.
Now yet another application has emerged: DNA data storage. It uses synthetic DNA molecules to encode and store digital information such as text or images. Binary code is thus translated into the genetic alphabet A, C, G and T, or the nucleotides Adenine, Cytosine, Guanine and Thymine.
Compact, long-term storage
Why would we use a biological molecule to do this work? Because it allows for an extremely high data density – one gram of DNA can potentially store petabytes of data. And it has an extremely long-term stability – thousands of years if stored under the right conditions. DNA storage might therefore help to solve the data storage crisis, with predicted data storage needs approaching 200 Zetabytes in 2025 and a YottaByte by 2030.
DNA data storage is especially interesting for long-term data archiving of e.g. (surveillance) video, bank transactions and medical and scientific data. Also, it shows clear advantages in terms of sustainability: a data storage server park as large as a football field can be replaced by a DNA storage server of merely a football in size.
Along with close to 50 other members, imec supports the DNA data storage alliance and co-authored its first whitepaper – Preserving our digital legacy: an introduction to DNA data storage – covering all the basics of DNA data storage.
How semiconductor technology addresses DNA storage challenges
Despite its great promises, DNA storage will have limited use in the immediate future. First, some hurdles will have to be overcome:
- slow reading and writing times
- high DNA synthesis and sequencing cost
- need for specialized equipment and personnel
To ensure a real breakthrough of DNA data storage, the cost of code writing and reading should go down. For the reading – or sequencing – we have seen a tremendous cost reduction since the start of the Human Genome Project in the early 1990s. Today, the cost is 30 USD per gigabase as announced early in 2023 on the leading genomics conference (AGBT).
More innovations are needed for low-cost sequencing/reading of DNA:
- very-high-density CMOS or CMOS imager based read-out arrays
- smart precision fluidics to enable addressability
- a quick sensor technology, such as solid-state nanopores
- integrated photonics to accelerate reading and error correction
- smart DNA encoding and sequence analysis software
Download the white paper ‘Semiconductor technologies and system concepts to revolutionize genomics’
Synthesizing DNA molecules could be done in a cost-effective way by using systems with high-density arrays of miniature electrodes on top of which numerous unique strands are grown simultaneously. To this end, imec develops:
- very-high-density micro-electrode arrays
- smart precision fluidics to enable addressability
- advanced CMOS electronics
Micro-electrode chip with 16,384 micro-electrodes
To truly enable a revolution in the world of fast DNA reading and writing, the two functionalities will need to be combined in one system to enable read and write of molecules without delay. Such a combined chiplet system with fast read and write would allow DNA storage to support a larger number of storage applications. It would even lead to a breakthrough in the solid-state memory market. To this end, imec’s expertise in advanced chiplet and 3D packaging concepts is extremely useful.
Work with us on DNA storage
Are you looking to accelerate your development of DNA storage solutions? Imec not only helps you to address technological challenges, but can also function as your full-fledged strategic partner – right down to venturing support.
Click the button below to get in touch.