Concentration-controlled Length-based Dna Computing for Weighted Graph Problems with Novel Readout Approach Using Real-time Pcr

DNA computing often makes use of annealing or hybridization, whether for vastly generating the initial candidate answers or amplification by using polymerase chain reaction. The main idea behind molecular search or DNA computing approaches for solving weighted graph problems is by controlling the degree of hybridization in order to generate more double stranded DNA, which represent the answer of the problem, during in vitro computation. Previously, length, concentration, and melting temperature, have been exploited to control the hybridization of DNA. In this research, we present an alternative length-based DNA computing approach, called directproportional length-based DNA computing, whereby the cost of each path is encoded by the length of the oligonucleotides in a proportional way. The initial pool generation can be done by utilizing hybridization-ligation method or parallel overlap assembly. After amplification is done by polymerase chain reaction, the result of the computation can be visualized by polyacrylamide gel electrophoresis, which is an operation to separate the respective double stranded DNA according to their length. The main advantage of the proposed method is that, after an initial pool generation and amplification, gel electrophoresis can be employed to directly decode the results of the computation. Further, we also proposed a hybrid approach, which is called concentration-controlled direct-proportional length-based DNA computing, that combines two characteristics: length and concentration of DNA. The length is crucial to encode the input data by oligonucleotides in a proportional way, as previous approach. On the other hand, the control of hybridization by means of concentration is incorporated by varying the amount of input oligonucleotides, with respect to the weights of input graph. The proposed hybrid approach is proved to be an effective method in terms of scalability and materials usage for solving weighted graph problems, such as the shortest path problem. In this research, the output of the shortest path problem is decoded by a method called graduated polymerase chain reaction. Lastly, a novel approach based on TaqMan real-time polymerase chain reaction for extraction of molecular information is also reported in this thesis. Normally, in graph problems based on DNA computing, such as Hamiltonian path problem, traveling salesman problem, and the shortest path problem, molecular information is normally required to in order to reveal the information encoded by DNA. It is found that based on the proposed novel approach, the molecular information can be extracted in less time compared to the conventional graduated polymerase chain reaction method. In addition, the performance of the proposed molecular extraction approach can be further upgraded by implementation of step-by-step analysis automatically on a silicon-based computer. This finding is a step forward towards the application of DNA computing in engineering field.