Cillian Bracken Conway
28th Apr, 2016

Microsoft is investing in 10 million strands of DNA for data storage, announces Twist Bioscience. With a single gram of DNA capable of storing close to 1 Zettabyte or 1 billion Terabytes of data, it makes sense to embrace the cutting-edge technology.

The DNA storage is developed by synthetic biology start-up Twist Bioscience, a company accelerating innovation and science through the rapid synthesis of high-quality DNA. According to the company CEO Emily M. Leproust, PhD, there are several advantages to DNA storage.

  • It has an almost infinite shelf life that spans several thousand years, as proven by a 7000-year-old DNA found in a cave.
  • It eliminates the need to re-encode data, as the storage medium doesn’t have a limited lifespan.
  • It offers a permanent storage format that can be read at continually decreasing costs.
  • It provides a long-term method for securing data and helps keep up with the exponentially expanding digital data.

“We could encode and recover 100% of the Digital Data from synthetic DNA”, Doug Carmean said, a Microsoft partner architect within Twist Bioscience’s Technology and Research organisation.

Microsoft has taken a chunk of data normally stored on a hard drive and translated it into genetic codes representative of a DNA’s chemical building block. These are As, Cs, Gs and Ts. The company then asked Twist bioscience to manufacture 10 million strands of DNA based on the sequence of codes they have provided.

The biggest challenge with DNA storage is reading and writing the codes, but the biology start-up has a machine that can produce custom strings of DNA. Microsoft only needs to supply the DNA sequence, and Twist Bioscience will make the DNA from scratch.

Only Microsoft has the decoder key for the sequence of codes they supplied to Twist, according to Leproust. The information is kept as a trade secret as well.

After the data has been transformed into invisible molecules in a test tube, the company will send it back to Microsoft for testing. There are ways to read the data from out of the test tubes. There are also methods to stimulate the passing of time by millennia.

The current cost of a custom DNA sequence is about $0.10 per base. Twist hopes to bring the cost down to $0.02. The cost of genetic sequencing used to read data has already dropped significantly, adding more benefits to using DNA storage.

The start-up company admits that it would be years before the DNA storage becomes commercially viable. But based on initial tests, the company is capable of substantially increasing “the density and durability of data storage”.

Microsoft also confirmed that initial trials allowed full retrieval of the data encoded in the DNA.