Author's School

Graduate School of Arts & Sciences

Author's Department/Program

Biology and Biomedical Sciences: Computational and Systems Biology

Language

English (en)

Date of Award

1-1-2011

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Gary Stormo

Abstract

The aim of this dissertation is two-fold:: 1) To catalog all cis-regulatory elements within the intergenic and intronic regions surrounding every gene in C.elegans: i.e. the regulome) and: 2) to determine which cis-regulatory elements are associated with expression under specific conditions. We initially use PhyloNet to predict conserved motifs with instances in about half of the protein-coding genes. This initial first step was valuable as it recovered some known elements and cis-regulatory modules. Yet the results had a lot of redundant motifs and sites, and the approach was not efficiently scalable to the entire regulome of C. elegans or other higher-order eukaryotes. Magma: Multiple Aligner of Genomic Multiple Alignments) overcomes these shortcomings by using efficient clustering and memory management algorithms. Additionally, it implements a fast greedy set-cover solution to significantly reduce redundant motifs. These differences make Magma ~70 times faster than PhyloNet and Magma-based predictions occur near ~99% of all C. elegans protein-coding genes. Furthermore, we show tractable scaling for higher-order eukaryotes with larger regulomes. Finally, we demonstrate that a Magma-predicted motif, which represents the binding specificity for HLH-30, plays a critical role in the host-defense to pathogenic infections. This novel finding shows that hlh-30(-) animals are more susceptible to S. aureus and P. aeruginosa than their wild type counterparts.

DOI

https://doi.org/10.7936/K7M906NX

Comments

Permanent URL: http://dx.doi.org/10.7936/K7M906NX

Share

COinS