Mar 17 2016
Long time readers are no doubt familiar with my facination with the subject of biological computing, using organic structures to process and represent information rather than silicon-hybrid substrates. When you get right down to it, DNA is an information storage and representation system, just like the tape upon which a notional Turing machine reads and writes symbols. Using this metaphor (which isn't nearly as tortured as it sounds), the ribosomes of eukaryotic cells would be the Turing machines that read the tape and carry out the operations (protein synthesis) encoded in the nucleotides.
Not too long ago the field advanced another crucial step. Rather than cutting-and-splicing long strings of DNA to hardcode programs into them, a research team at MIT figured out a way to represent basic operations with much smaller segments of DNA. Bacteria exchange plasmids, short loops of DNA that encode single genes which are used to mix up the gene pool of a species of bacteria in close quarters. Timothy Lu, who lead the research team, designed a set of synthetic plasmids for escherichia coli, which is one of the most commonly encountered (and studied) bacteria by humans. Each plasmid represents a basic logic or arithmatic function, from addition to exclusive-or. In addition to promoter and terminator nucleotide sequences (which mark the beginning and end of the gene in the plasmid) the payload of the nucleotide codes for a protein which glows green. Recombinase enzymes that cut and splice DNA at specific loci are used to encode inputs by splicing and arranging plasmids into longer sequences of DNA, turning them into programs comprised of smaller operations. The DNA is assembled within and expressed by the bacterial culture, and ultimately the bacteria either do or do not produce the fluorescent protein (effectively outputting a 1 or a 0).
So, what does this ultimately mean? What good is either producing or not producing a protein that glows under UV light. Seems kind of pointless.
Looking at it from the 30,000 foot view, what we have here is a way to feed new and potentially novel gene sequences into one of the most common bacteria out there in a more reliable way. It is also now possible to assemble them in a more symbolic fashion, with boolean logic and arithmatic rather than by reverse engineering existing proteins and figuring out which patterns of nucleotides correspond to which amino acids in an organism. It is also now possible to turn those gene sequences on and off after they've been assimilated by the bacterial culture at a later time. This means that synthesis of compounds could be started and stopped without killing the culture at a later time - it is plausible that the cell culture could halt synthesis of whatever it's been programmed for when it reaches a certain concentration or internal threshold.