Dissertation/Thesis Abstract

The Detection and Characterization of Epistasis and Heterogeneity: A Learning Classifier System Approach
by Urbanowicz, Ryan John, Ph.D., Dartmouth College, 2012, 288; 3507841
Abstract (Summary)

As the ubiquitous complexity of common disease has become apparent, so has the need for novel tools and strategies that can accommodate complex patterns of association. In particular, the analytic challenges posed by the phenomena known as epistasis and heterogeneity have been largely ignored due to the inherent difficulty of approaching multifactor non-linear relationships. The term epistasis refers to an interaction effect between factors contributing to diesease risk. Heterogeneity refers to the occurrence of similar or identical phenotypes by means of independent contributing factors. Here we focus on the concurrent occurrence of these phenomena as they are likely to appear in studies of common complex disease. In order to address the unique demands of heterogeneity we break from the traditional paradigm of epidemiological modeling wherein the objective is the identification of a single best model describing factors contributing to disease risk. Here we develop, evaluate, and apply a learning classifier system (LCS) algorithm to the identification, modeling, and characterization of susceptibility factors in the concurrent presence of heterogeneity and epistasis. This work includes an examination of existing LCS algorithms, the development of new strategies to address the inherent problem of knowledge discovery in LCSs, the introduction of novel heuristics that improve LCS performance and allow for the explicit characterization of heterogeneity, and an application of these collective advancements to an investigation of bladder cancer susceptibility. Successful analysis of simulated and real-world complex disease associations demonstrates the validity and unique potential of this LCS approach.

Indexing (document details)
Advisor: Moore, Jason H.
Commitee: Eppstein, Margaret J., Gross, Robert H., Thornton-Wells, Tricia A., Whitfield, Michael L.
School: Dartmouth College
Department: Genetics
School Location: United States -- New Hampshire
Source: DAI-B 73/09(E), Dissertation Abstracts International
Subjects: Genetics, Bioinformatics, Computer science
Keywords: Data mining, Epistasis, Evolutionary algorithm, Genetic association study, Heterogeneity, Learning classifier system
Publication Number: 3507841
ISBN: 978-1-267-33318-6
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy