Humanity has a huge data storage problem. In the past two years, people have created more information than in all of previous history. And this stream of information will soon surpass the capacity of hard drives.
The researchers say they have found a new way to encode digital information in DNA. One gram of DNA can store 215 petabytes (215 million gigabytes) of data. Thus, all information ever created by man will occupy a container the size of a couple of trucks.
DNA has many advantages for storing digital information. It is ultra compact and can be stored for thousands of years in a cool, dry place. And people can always decipher it. “DNA does not degrade over time like cassettes or discs, and it does not become outdated,” says Yaniv Ehrlich, a scientist at Columbia University (USA).
Scientists have been preserving digital information in DNA since 2012, when geneticists from Harvard University (USA) George Church, Sri Kosuri and their colleagues encrypted a book of 52 thousand words into thousands of DNA fragments using strands from the four-letter alphabet - A, G, T and C to encode the zeros and ones of the digitized file.
This encryption system was relatively ineffective and could only store 1.28 petabytes per gram of DNA. Other approaches have worked better. But none allowed DNA to retain more than half of its maximum capacity. DNA can withstand about 1.8 bits per DNA nucleotide (the number does not reach 2 bits due to rare but unavoidable read and write errors).
Ehrlich decided that he would come closer to this limit. Therefore, he and Dina Zilinski turned to the algorithms that were used to encrypt and decrypt information. They started with 6 files, which included a complete computer operating system, a computer virus, an 1895 French film titled Arrival of a Train at La Ciotat, and a 1948 study by theorist Claude Shannon. First, the scientists converted the files into binary strings of ones and zeros, compressed them into one base file, and then split the data into short strings of binary code. They developed an algorithm called the "Fountain of DNA", which randomly packs chains into so-called "blobs". The researchers added additional tags to these so that they could be rebuilt later in the correct order. In total, scientists have generated a digital list of 72 thousand DNA strands,each 200 characters in length.
Promotional video:
They sent them as text files to the Twist Bioscience startup in California, where they synthesized DNA strands. Two weeks later, Ehrlich and Zilinski received an ampoule with a piece of DNA in the mail, in which their files were encrypted. To decipher them, scientists used modern DNA sequencing technology. The sequences were sent to a computer, which translated the genetic code back into binary and used tags to reassemble the six original files. The technology worked so well that the new files were free of errors.
However, Kosuri and Ehrlich noted that the new approach is not ready for large-scale use. They spent 7 thousand dollars to synthesize 2 megabytes of information into files, and another 2 thousand dollars to read it. Compared to other forms of data storage, writing and reading from DNA is relatively slow.