Clustering techniques for DNA computing readout method based on real-time polymerase chain reaction

Abstract

In the first experiment of Deoxyribonucleic Acid (DNA) computation, Adleman has solved a seven nodes Hamiltonian Path Problem (HPP) by applying some biotechnology techniques such as hybridization and polymerase chain reaction (PCR). In that experiment, graduated PCR has been used to visualize the Hamiltonian path. In other research work, a novel readout method tailored specifically to the HPP in DNA computing was proposed, which employs a hybrid in vitro-in silico approach. In the in vitro phase, TaqMan-based real-time PCR reactions are performed in parallel, to investigate the ordering of pairs of nodes in the Hamiltonian path, in terms of relative distance from the DNA sequence encoding the known start node. The resulting relative orderings are then processed in silico, which efficiently returns the complete Hamiltonian path. However, this method used manual classification to distinguish the two different reactions of real-time PCR. In this thesis, clustering techniques are implemented during the in silico phase. Clustering is crucial to identify automatically two different reactions produced by real-time PCR. K-means, Fuzzy C-means (FCM), and Alternative Fuzzy C-means (AFCM) clustering algorithms are implemented to differentiate the output of realtime PCR. Results show that K-means and FCM clustering algorithms are capable to classify the two different reactions of real-time PCR. In addition, it has been shown that AFCM clustering algorithm is better than FCM and K-means in term of handling outliers in the real-time PCR output data. Application of clustering techniques have improved the in silico information processing of the readout method.