Knowledge discovery and modeling in genomic databases
Department of Computer and Information Science
Doctor of Philosophy
Computer and Information Science
Wang, Jason T. L.
McHugh, James A.
Shih, Frank Y.
Hidden Markov models
This dissertation research is targeted toward developing effective and accurate methods for identifying gene structures in the genomes of high eukaryotes, such as vertebrate organisms. Several effective hidden Markov models (HMMs) are developed to represent the consensus and degeneracy features of the functional sites including protein-translation start sites, mRNA splicing junction donor and acceptor sites in vertebrate genes. The HMM system based on the developed models is fully trained using an expectation maximization (EM) algorithm and the system performance is evaluated using a 10-way cross-validation method. Experimental results show that the proposed HMM system achieves high sensitivity and specificity in detecting the functional sites.
This HMM system is then incorporated into a new gene detection system, called GeneScout. The main hypothesis is that, given a vertebrate genomic DNA sequence S, it is always possible to construct a directed acyclic graph G such that the path for the actual coding region of S is in the set of all paths on G. Thus, the gene detection problem is reduced to the analysis of paths in the graph G. A dynamic programming algorithm is employed by GeneScout to find the optimal path in G. Experimental results on the standard test dataset collected by Burset and Guigo indicate that GeneScout is comparable to existing gene discovery tools and complements the widely used GenScan system.
njit-etd2002-084 (128 pages ~ 4,917 KB pdf)
Please complete this Feedback Form to inform us about your experience using this website. It will assist us in better serving your information needs in the future. Thank You!
Created August 12, 2003