Date of Award

Spring 2006

Document Type


Degree Name

Doctor of Philosophy (PhD)


Computational Analysis and Modeling

First Advisor

Chokchai Leangsuksun


The healthcare burden and suffering due to life-threatening diseases such as cancer would be significantly reduced by the design and refinement of computational interpretation of micro-molecular data collected by bioinformaticians. Rapid technological advancements in the field of microarray analysis, an important component in the design of in-silico molecular medicine methods, have generated enormous amounts of such data, a trend that has been increasing exponentially over the last few years. However, the analysis and handling of these data has become one of the major bottlenecks in the utilization of the technology. The rate of collection of these data has far surpassed our ability to analyze the data for novel, non-trivial, and important knowledge. The high-performance computing platform, and algorithms that utilize its embedded computing capacity, has emerged as a leading technology that can handle such data-intensive knowledge discovery applications.

In this dissertation, we present a novel framework to achieve fast, robust, and accurate (biologically-significant) multi-class classification of gene expression data using distributed knowledge discovery and integration computational routines, specifically for cancer genomics applications. The research presents a unique computational paradigm for the rapid, accurate, and efficient selection of relevant marker genes, while providing parametric controls to ensure flexibility of its application.

The proposed paradigm consists of the following key computational steps: (a) preprocess, normalize the gene expression data; (b) discretize the data for knowledge mining application; (c) partition the data using two proposed methods: partitioning with overlapped windows and adaptive selection; (d) perform knowledge discovery on the partitioned data-spaces for association rule discovery; (e) integrate association rules from partitioned data and knowledge spaces on distributed processor nodes using a novel knowledge integration algorithm; and (f) post-analysis and functional elucidation of the discovered gene rule sets. The framework is implemented on a shared-memory multiprocessor supercomputing environment, and several experimental results are demonstrated to evaluate the algorithms. We conclude with a functional interpretation of the computational discovery routines for enhanced biological physiological discovery from cancer genomics datasets, while suggesting some directions for future research.