Hybrid Genetic Approach to Oligo Sets Optimization Piotr W¸ asiewicz, Grzegorz Tomczuk, Jan J. Mulawka Institute of Electronic Systems Warsaw University of Technology Nowowiejska 15/19, 00-665 Warsaw, Poland {pwasiewi, gtomczuk}@elka.pw.edu.pl {pwas,jml}@ise.pw.edu.pl Abstract. DNA computing is a striking new information technology based on chemical reactions in tubes utilizing specially designed with a help of computer programs DNA polymers. This methodology provides new molecular mechanism for storing and processing information. DNA macrostructures are bases of specially designed algorithms realized by so called soft hardware applications. To obtain these structures a special DNA sequences design tool is required. A custom genetic algorithm with new hybrid operators was involved in creating a set of DNA strings. Changes in the input files and examples of string generation were introduced. 1 Introduction In the new century a great progress in the area of nanoelectronics is expected. Traditional methodologies will soon meet their technological limits. Possibilities of silicon electronic chips further miniaturization are almost exhausted. Investigations indicate that different other chemical compounds can be used to store and process information on molecular scale. Especially organic substances used in biology are well suited for this purpose e.g. nucleic acids and proteins. One of first steps in this direction was developed by cooperation of computer scientists and genetic engineers. As a result information technology based on computing utilizing molecules during chemical reactions has appeared. This new methodology is called DNA computing. A single-strand DNA has a phospho-sugar backbone with two different, 5’ and 3’ ends and four bases Adenine, Thymine, Cytosine, Guanine denoted by the symbols A, T, C, and G, respectively. A double-strand DNA may be formed of oriented in the opposite directions two single strings due to hybridization or in other words annealing reaction, because A is complementary with T, and C is complementary with G. Due to this reaction the oligonucleotides may connect with each other during concatenation process called ligation forming longer DNA chains [17]. A sequence of such operations on DNA strings is called an algorithm. But in the typical DNA computing algorithm this sequence is determined by a model of DNA strings similar to the soft hardware specialized architecture driven here by heating, cooling and connected with them operations on DNA. Together the operation sequence and the model make computation possible. First DNA computing method was invented by Adleman [9]. He demonstrated how to solve NP-complete combinatorial and graph problem of finding the Hamilton path that is very difficult to solve for conventional computers. His work began in this field further research, which was reported in many papers describing new DNA computing applica- tions [1–15]. It has been demonstrated that DNA computing is suitable for programming in logic [10, 11] and for solving NP-complete problems [10]. For example DNA computing molecular inference systems [2] are a first step towards creation of the fifth computer hardware generation based on logic and the Prolog language structure. And multidimen- sional DNA computing leads in future to applications of molecular electronics in at least