We sometimes mistake our computers for living organisms. Now it's the turn of living organisms to become computers. This is the project of a team of Harvard researchers: to use bacteria in such a way that they are able to store our data, more efficiently than any hard drive would do.
Uteam of Harvard scientists, led by geneticists Seth Shipman and Jeff Nivala, have just developed a method for writing information into the genetic code of living things. In this case it involves bacteria... e-Coli that become the living medium of non-living information: a passage of a poem or a computer program, it doesn't matter. What is even more remarkable is that this information brought to the bacterium is transmitted hereditarily to the descendants of the carrier bacteria. Scientists then simply read the data by genotyping the bacteria. The article by the authors of this experiment published in the journal Science says that the method can now store 100 bytes of data but that capacity will grow very quickly.
We wanted to know if we could use nature's own methods to write directly on the genome... "
How did the researchers achieve this feat? By drawing inspiration from the now famous genetic editing tool, CRISPR. In nature, when a bacterium is attacked by a virus, it deploys a very sophisticated immune defence: the attacked bacterium will cut out a segment of DNA from the virus and stick it in a specific region of its own genome. It proceeds as if it had natural CISPR scissors at its disposal. This allows the bacteria to retrieve the virus' "signature" so that it can remember it if it attacks it again. This genetic memory is not only retained but also passed on to the bacteria's descendants.
The team of geneticists at Harvard found that if a segment of genetic data was introduced that looks like to viral DNA in a colony of bacteria, these little bugs will use their natural CRISPR to incorporate the data into their genetic code. So, to transform a colony of bacteria e-Coli in as many hard drives, Shipman and Nivala scattered free segments of fake viral DNA throughout the colony.
In their experiment, the researchers introduced arbitrary strings of data in the form of messages written with the four main letters of the genetic alphabet of life: A, T, C, G. Once this data is introduced, the bacteria do the rest of the work and store the information like perfect librarians.
One important detail has not escaped researchers: bacteria store new immune system memories sequentially. This means that the data extracted from viral DNA is arranged in chronological order of their appearance in the life of the bacteria. This feature is fundamental for Steh Shipman, who told the scientific journal Popular Mechanics : " If the new information was just randomly stored, it wouldn't be as informative. Labels would have to be placed on each piece of information to show when it was entered into the cell. Here, it is ordered sequentially, the same way you organize words in a sentence ".
So certainly not all bacteria in a colony will pick up the coded message. Some miss it. If the message was a sentence, not all the words in the sentence would be found in all the bacteria. But that's not a real problem for researchers. Indeed, Shipman explains that one can very quickly genotype millions of bacteria in a colony to statistically extract the entire message, without any loss of information.
The experiment conducted by the Harvard team focuses on the bacterium e-Coli which can only support 100 bytes of data. But other microorganisms such as Sulfolobus tokodaii would already be able to store 3000 bytes of data. Researchers say that with synthetic engineering, it is easy to imagine bacteria designed specifically to act as hard drives, with expanded regions of storage in their genetic code. This will then open the way to downloading and storing very large amounts of data. Data that is indestructible because it is faithfully replicated over generations of carrier organisms.
