Inference Models in DNA Computing

Biomolecular Computation is a discipline that deals with the design and implementation of processing devices using a biological substrate. Due to the stability of the DNA molecule and the ease to predict the interactions between DNA strands, it has become one of the most widely used substrates, to the extend that the discipline is often related as DNA Computing. After the early models that mainly focused on the resolution of NP-complete problems, it was understood that biomolecular computers could not compete in speed and accuracy with silicon based computers. Therefore the objectives of the discipline changed, moving away from the resolution of classical computing problems towards problems where biomolecular computers show their strength with in vitro and in vivo biomedical applications. The talk will present different models of DNA devices able to implement various types of inference (logical or probabilistic) exploiting different biomolecular computation techniques (DNA strand displacement and enzymatic reactions) with potential applications in genetic diagnosis.