Microsoft automatically writes and reads ‘hello’ in DNA with a single system

Microsoft has reduced the techniques for storing and reading data from DNA to a single automated system. In a demonstration, the system described ‘hello’, or 5 bytes, in DNA.

First, Microsoft and research partner the University of Washington use software to convert bits of a file into synthetic DNA strands of the four different nucleotides a, t, c and g. Relatively simple lab equipment and chemicals are used to bond the strands to a solid for storage.

Chemicals in a liquid allow a microfluidic pump to transport the strands to a reader. That can convert the DNA back into bits that can be read by a computer. In total, the system costs about ten thousand dollars and storage and reading takes about 21 hours thanks to slow chemical reactions.

Speed ​​or low cost were not a goal of the study, according to Microsoft. However, the set-up is intended to demonstrate that writing, storing and reading can be automated, without technicians having to perform all kinds of steps in the laboratory.

Microsoft and the Washington research team are also working to automate the microfluidic arrangement of droplets, combined with software. This combination of PurpleDrop hardware and Puddle software should provide speed improvements and could also help in the medical world for diagnostics, for example.

Researchers from Microsoft’s Molecular Information Systems Lab and the University of Washington have published their research in Nature under the title Demonstration of End-to-End Automation of DNA Data Storage.

Microsoft has been experimenting with DNA storage for years. In 2016, the company managed to store and read 200MB of DNA, including Project Gutenberg’s top 100 books.