We recently reported on a new development wherein researchers were able to successfully encode data onto DNA, simulate 2000 years of storage and then decode it off the DNA without any errors. This development sparked a frenzy wherein it was suggested that DNA will be replacing hard drive, but this isn’t going to happen in near future and if it does, it won’t be reaching first.
The reason? Cost! Robert Grass, Ph.D., at ETH Zurich revealed through their study that they successfully encoded 83 kilobytes of text onto DNA, then encapsulated the encoded DNA in silica spheres and warmed it to nearly 160 degrees Fahrenheit for one week, which is the equivalent of keeping it for 2,000 years at about 50 degrees and when they decoded it, it was error-free.
Grass does note that though they successfully managed to pull it off, the cost was somewhere around $2,000. Storage of just a few megabytes of data onto DNA could cost thousands of dollars – something that even enterprises would think about at least twice!
Then there is the problem of indexing. What if you want to read a specific file from the whole bunch of files encoded in the DNA? This is another unsolved issue that Grass and his team are trying to find a solution for. Grass notes that they are able to read everything stored on the DNA, but there is no way of reading a specific set of data located at a particular system. In nutshell, researchers have still to come up with a file system for DNA storage with efficient indexing.
The use of specialised equipment is another challenge that needs to be overcome. Beyond the cost of encoding and decoding data onto DNA, the instruments required to pull if off are not commercially available.
Further, what about the time it takes to encode and decode data? According to an article published in Nature Communications, in June it takes days to encode just 10MB of data and about eight hours to decode that same 10MB. Forget about whole length movies, even if you have a multimedia presentation that needs to be encoded it would take you days before you can store it.
Researchers expect that they will be able to encode a feature-length movies in about two to three years, at which time the technology will begin to become commercially viable.