Splicing systems for universal turing machines

  • Authors:
  • Tero Harju;Maurice Margenstern

  • Affiliations:
  • Department of Mathematics, University of Turku, Turku, Finland;LITA, EA3097, UFR MIM, Université de Metz, Metz, France

  • Venue:
  • DNA'04 Proceedings of the 10th international conference on DNA computing
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we look at extended splicing systems (i.e., H systems) in order to find how small such a system can be in order to generate a recursively enumerable language. It turns out that starting from a Turing machine M with alphabet A and finite set of states Q which generates a given recursively enumerable language L, we need around 2 × |I| + 2 rules in order to define an extended H system ${\cal H}$ which generates L, where I is the set of instructions of Turing machine M. Next, coding the states of Q and the non-terminal symbols of ${\cal L}$, we obtain an extended H system ${\cal H}_1$ which generates L using |A| + 2 symbols. At last, by encoding the alphabet, we obtain a splicing system ${\cal U}$ which generates a universal recursively enumerable set using only two letters.