The DNA computer: super hard drive of the future?

Liv­ing com­puters are the prom­ise of DNA com­put­ing. This tech­no­logy pro­poses to exploit the com­pu­ta­tion­al cap­ab­il­it­ies of cells, repla­cing sil­ic­on sys­tems with bio­tech­no­logy. Although the approach is still in its infancy, it is already arous­ing a mix­ture of won­der and fear.

There are the four sym­bols, based on the nuc­le­otide bases, which are the molecules that con­sti­tute the DNA alpha­bet. A for aden­ine, T for thym­ine, C for cytosine and G for guan­ine. So, just as we code in bin­ary (0, 1) by play­ing on the dif­fer­ence in elec­tric­al charge, we can encode inform­a­tion in A, T, C and G. In prin­ciple, the concept not par­tic­u­larly com­plic­ated to do. The proof: in 2020, Twist Bios­cience, an Amer­ic­an com­pany spe­cial­ising in DNA syn­thes­is, Net­flix and the Zurich Poly­tech­nic announced that they had encoded the entire Ger­man series Bio­hack­ers – which centres its plot on the world of syn­thet­ic bio­logy – on only a few strands of DNA. A cut­ting-edge com­mu­nic­a­tions coup!

In fact, in terms of stor­age, DNA com­puters are much bet­ter than elec­tron­ic sys­tems. DNA stores inform­a­tion for thou­sands, even mil­lions of years without any energy input. Fossil DNA, for example, has shown us this. So, in that sense, it is a more sus­tain­able solu­tion, with great­er inform­a­tion dens­ity and bio­de­grad­ab­il­ity. But the tech­no­logy will not stop there. With an enzyme that can edit bases, a DNA sequence and cel­lu­lar machinery that reads DNA and pro­duces bio­lo­gic­al molecules, we already have a DNA com­puter. The real chal­lenge is the archi­tec­ture of such sys­tems. Lest we for­get John von Neu­mann, the Hun­gari­an-Amer­ic­an math­em­atician who enabled mod­ern com­put­ing to emerge by invent­ing an archi­tec­ture enabling pro­cessors to com­mu­nic­ate with ran­dom access memory (RAM) when he was a con­sult­ant at IBM. This is what DNA com­puters need today: an architecture.

DNA molecules are memor­ies. As far as pro­cessors are con­cerned, recent work has high­lighted ‘tran­script hubs’ in stem cells, i.e. regions con­cen­trat­ing a large num­ber of sys­tems for read­ing and con­trolling DNA machinery. These regions are linked to the organ­isa­tion of DNA in space, to the way in which dis­tant sequences of the gen­ome can come togeth­er when the DNA strand folds. They are like the micro­pro­cessors in DNA com­puters, which can con­trol both access to and the read­ing of dif­fer­ent regions of the gen­ome. Thanks to this dis­cov­ery, we can ima­gine that with­in 10 or 20 years, labor­at­or­ies will have real demon­strat­ors of bio­lo­gic­al computers.

DNA is a bio­de­grad­able mater­i­al. How­ever, DNA com­puters will nev­er entirely replace sil­ic­on chips. We won’t see DNA tele­phones, for example. This is because to read inform­a­tion in DNA with cur­rent sequen­cing tech­niques we must extract the genet­ic mater­i­al from the sample and there­fore des­troy the chip or cell that con­tained it. What’s more, to encode the inform­a­tion requires syn­thet­ic bio­logy, which is time-con­sum­ing, costly, and dif­fi­cult to gen­er­ate long frag­ments. A great deal of research is needed to improve these processes.

In any case, the research that needs to be car­ried out to ima­gine DNA com­puters will be very instruct­ive about how liv­ing things work. In this sense, even if this tech­no­logy nev­er comes to fruition, it will advance our under­stand­ing of life. In the same way that the steam engine made it pos­sible to mas­ter the laws of ther­mo­dy­nam­ics, DNA com­puters will help us to under­stand the ther­mo­dy­nam­ics of life.

Pro­gram­ming a cell is a way of talk­ing to a life form. By pro­du­cing trans­miss­ible trans­form­a­tions in these DNA com­puters, we will be able to com­mu­nic­ate with liv­ing beings over sev­er­al gen­er­a­tions. DNA is a uni­ver­sal lan­guage between spe­cies on Earth. In this sense, isn’t read­ing the DNA of an extinct spe­cies a form of com­mu­nic­a­tion, even if it is one-way? How­ever, the lan­guage of liv­ing things is not like a human lan­guage. We shouldn’t hold out too much hope that we can crack jokes with trees or talk to a woolly mam­moth as if we were talk­ing to our dog.

DNA com­put­ing has the poten­tial, yes. But is it accept­able? Just because we have the power to do some­thing doesn’t mean we should. Pur­pose is a bioeth­ic­al issue that must not be ignored while we devel­op this tech­no­logy. And this ques­tion is all the more import­ant because most DNA com­puters will not be vis­ible. It will be hid­den inside cells, sim­il­ar to those that make up our bod­ies. Only spe­cial­ists will recog­nise the pres­ence of DNA computers. 

Agnès Vernet