That DNA is the carrier of our genetic information has been known ever since the 1950s. In recent years, however, it has become more and more obvious that this amazing molecule may also be a handy tool for engineers. Indeed, on the basis of DNA you can even make some kind of computers.

Applications of DNA computers, as Professor of Protein Engineering Maarten Merkx (on the right in the photo) explains, can be anticipated especially in the biomedical field. That is because this biomolecule feels like a fish in water in living systems. When you think of DNA computers you should not think of a buzzing box on top of your desk, but rather of molecules floating around, for example in the bloodstream, which engage in a chemical reaction when they meet. The new molecules that are formed in these reactions can in their turn start yet another reaction: the next step in the ‘computation’.

In a test tube this turns out to work exceptionally well already, says Merkx. “The specific DNA molecules that you need for a calculation may simply be ordered online. For some twenty odd euros they are delivered at your doorstep within two days, and then they actually always do what they are supposed to do.” However, for anything useful is to be done with such a computation, the DNA computer also needs an ‘interface’ with its environment: you need to be able to provide specific input and also, afterwards, to be able to read out or use the result of the computation.

Recently Merkx along with his PhD candidate Wouter Engelen and colleagues from Computational Biology have published an article in the Nature Communications journal, in which they show that they can use antibodies – molecules that we make against pathogens - as input for a DNA computer.

The new method, which the Eindhoven researchers have christened Antibody-Templated Strand Exchange, has been put into a mathematical model by Lenny Meijer and Tom de Greef (on the left in the photo) from Computational Biology. By means of this model they managed to predict with which molecules and concentrations the computer would function optimally.