It is estimated that digital data will reach 44 trillion gigabytes by 2020, a tenfold increase from figures in 2013. That's enough to create six stacks of computer tablets reaching the moon! Will there be enough storage? Researchers from Microsoft and the University of Washington (UW) are looking to address this issue by turning to DNA as a storage solution.

In a study, the researchers detailed a new technique they have developed which allowed them to successfully encode four image files worth of digital data into the nucleotide sequences of snippets of synthetic DNA.

More importantly, they were able to reverse the process and retrieve the right sequences representing the encoded digital data from a larger DNA pool and reconstruct the four images without compromising an information byte.

"We're essentially repurposing [DNA] to store digital data ... in a manageable way for hundreds or thousands of years," said Luis Ceze, co-author for the study and UW computer science and engineering associate professor.

Molecules of DNA are capable of storing information more densely millions of times than current storage technologies, which also deteriorate after several years. DNA, on the other hand, not only has an infinitely larger storage capacity but can also last for centuries, making it perfect for archival purposes.

Currently, the biggest challenge to using DNA as a storage solution is the cost and efficiency associated with synthesizing and sequencing DNA on a larger scale. With the right incentives in place, however, the researchers are confident that any technical barrier can be overcome.

Using DNA as storage mostly relies on biotechnology techniques, but the practice can incorporate new expertise as well. For the current study, the researchers took advantage of error correction schemes that have not been used on DNA before, just in computer memory.

Other authors for the study [pdf] include: Karin Srauss, Georg Seelig, Douglas Carmean, Randolph Lopez and James Bornholt. They received funding support from the David Notkin Endowed Graduate Fellowship, the National Science Foundation and Microsoft Research.

ⓒ 2021 All rights reserved. Do not reproduce without permission.