DNA computing based on splicing: universality results

Abstract

First, we recall some characterizations of recursively enumerable languages by means of finite H systems with certain regulations on the splicing operation. Then, we consider a variant of the splicing operation where the splicing proceeds always in couples of steps: the two strings obtained after a splicing enter immediately a second splicing (the rules used in the two steps are not prescribed). Somewhat surprising if we take into account the loose control on the performed operations, extended H systems with finite sets of axioms and of splicing rules, using this double splicing operation, can again characterize the recursively enumerable languages. Finally, we consider two types of distributed H systems: communicating distributed H systems and time-varying distributed H systems. For the first type of devices, we give a new proof of the recent result of [25] that (in the extended case) such systems with three components characterize the recursively enumerable languages. In what concerns the second mentioned distributed model, we prove that time-varying H systems with seven components can characterize the recursively enumerable languages. The optimality of these two last-mentioned results is open.