Mathuri Selvarajoo

Probabilistic Splicing Systems

In this paper we introduce splicing systems with probabilities, i.e., probabilistic splicing systems, and establish basic properties of language families generated by this type of splicing systems. We show that a simple extension of splicing systems with probabilities may increase the computational power of splicing systems with finite components.

Probabilistic simple sticker systems

A model for DNA computing using the recombination behavior of DNA molecules, known as a sticker system, was introduced by by L. Kari, G. Paun, G. Rozenberg, A. Salomaa, and S. Yu in the paper entitled DNA computing, sticker systems and universality from the journal of Acta Informatica vol. 35, pp. 401-420 in the year 1998. A sticker system uses the Watson-Crick complementary feature of DNA molecules: starting from the incomplete double stranded sequences, and iteratively using sticking operations until a complete double stranded sequence is obtained. It is known that sticker systems with finite sets of axioms and sticker rules generate only regular languages. Hence, different types of restrictions have been considered to increase the computational power of sticker systems. Recently, a variant of restricted sticker systems, called probabilistic sticker systems, has been introduced [4]. In this variant, the probabilities are initially associated with the axioms, and the probability of a generated string is ...

The properties of probabilistic simple regular sticker system

A mathematical model for DNA computing using the recombination behavior of DNA molecules, known as a sticker system, has been introduced in 1998. In sticker system, the sticker operation is based on the Watson-Crick complementary feature of DNA molecules. The computation of sticker system starts from an incomplete double-stranded sequence. Then by iterative sticking operations, a complete double-stranded sequence is obtained. It is known that sticker systems with finite sets of axioms and sticker rule (including the simple regular sticker system) generate only regular languages. Hence, different types of restrictions have been considered to increase the computational power of the languages generated by the sticker systems. In this paper, we study the properties of probabilistic simple regular sticker systems. In this variant of sticker system, probabilities are associated with the axioms, and the probability of a generated string is computed by multiplying the probabilities of all occurrences of the initial strings. The language are selected according to some probabilistic requirements. We prove that the probabilistic enhancement increases the computational power of simple regular sticker systems.

Probabilistic simple splicing systems

A splicing system, one of the early theoretical models for DNA computing was introduced by Head in 1987. Splicing systems are based on the splicing operation which, informally, cuts two strings of DNA molecules at the specific recognition sites and attaches the prefix of the first string to the suffix of the second string, and the prefix of the second string to the suffix of the first string, thus yielding the new strings. For a specific type of splicing systems, namely the simple splicing systems, the recognition sites are the same for both strings of DNA molecules. It is known that splicing systems with finite sets of axioms and splicing rules only generate regular languages. Hence, different types of restrictions have been considered for splicing systems in order to increase their computational power. Recently, probabilistic splicing systems have been introduced where the probabilities are initially associated with the axioms, and the probabilities of the generated strings are computed from the probabilities of the initial strings. In this paper, some properties of probabilistic simple splicing systems are investigated. We prove that probabilistic simple splicing systems can also increase the computational power of the splicing languages generated.

Some characteristics of probabilistic one-sided splicing systems

A theoretical model for DNA computing using the recombination behavior of DNA molecules known as asplicing system has been introduced in 1987. Splicing systems are based on the splicing operation which, informally, cuts two strings at the specific places and attaches the prefix of the first string to the suffix of the second string and the prefix of the second string to the suffix of the first string yielding the new strings. It is known that splicing systems with finite sets of axioms and splicing rules only generate regular languages. Hence, different types of restrictions for splicing systems have been considered to increase the computational power of the languages generated. Recently, probabilistic splicing systems have been introduced where the probabilities are initially associated with the axioms, and the probabilities of the generated strings are computed from the probabilities of the initial strings. In this paper, some properties of probabilistic one-sided splicing systems, which are special types of probabilistic splicing systems, are investigated. We prove that probabilistic one-sided splicing systems can also increase the computational power of the languages generated.