Adaptive Double Self-Organizing Map for Clustering Gene Expression Data

In presenting this thesis in partial hlfillment of the requirements for an advanced degree at The University of Maine, I agree that the Library shall make it freely available for inspection. I further agree that the Librarian may grant permission for "fair use" copying of this thesis for scholarly purposes. It is understood that any copying or publication of this thesis for financial gain shall not be allowed without my written permission. This thesis presents a novel clustering technique known as adaptive double self-organizing map (ADSOM) that addresses the issue of identifying the "correct" number of clusters. ADSOM has a flexible topology and performs clustering and cluster visualization simultaneously, thereby requiring no a priori knowledge about the number of clusters. ADSOM combines features of the popular self-organizing map with two-dimensional position vectors, which serve as a visualization tool to decide the number of clusters. It updates its free parameters during training and it allows convergence of its position vectors to a fairly consistent number of clusters provided that its initial number of nodes is greater than the expected number of clusters. A novel index is introduced based on hierarchical clustering of the final locations of position vectors. The index allows automated detection of the number of clusters, thereby reducing human error that could be incurred from counting clusters visually. The reliance of ADSOM in identifjmg the number of clusters is proven by applying it to publicly available gene expression data from multiple biological systems such as yeast, human, mouse, and bacteria. providing me with the opportunity to pursue my Master's degree at University of Maine. I would like to thank them for all their time, encouragement and guidance during my over two-year graduate study. I would also like to thank Dr. Cristian Dornnisoru for his kindly assistance in my graduate research work as well as in many other things at the University of Maine. I wish to thank graduate coordinator Dr. all of their time and assistance with the various courses I have taken. Thanks also go to Ms. Padma Natarajan for her care and encouragement. I want to thank all the other faculty members in the department and other lab members in the Intelligent Systems Laboratory who have given me help during my graduate study at University of Maine. I would like to thank Dr. Su of the National Central University in Taiwan for his initial idea about double …