Abstract
The turn of the 21st century has been marked by a resurgence of interest in achieving a systems-level understanding of biology. One of the long-term goals of a newly emerged research field, called systems biology, aims at a comprehensive understanding of the genetic regulatory networks: how the genes dynamically interact and regulate each other to form highly coherent and coordinated physiological systems during the organism's development and in response to homeostatic challenges. DNA microarray and ChIP-chip are two high-throughput biotechnologies that provide complementary information of genetic regulatory networks. DNA microarray can simultaneously measure the mRNA level of each gene in a genome at a specific time point of a biological process being studied. ChIP-chip can simultaneously determine what are the target genes that a transcription factor (TF) may bind and what are the TFs that may bind to a gene. In this thesis, we integrate DNA microarray and ChIP-chip data to study the genetic regulatory networks of the yeast. First, we develop a computational method, called Temporal Relationship Identification Algorithm (TRIA), which uses DNA microarray data to identify a TF’s regulatory targets from its binding targets inferred from ChIP-chip data. TRIA can be thought of as a “refinement” process of ChIP-chip data to filter out the binding but non-regulatory targets of a TF. Finding the regulatory targets of TFs is important for understanding gene regulation. TRIA is helpful for achieving this purpose. Second, we develop a computational method, called MOdule Finding Algorithm (MOFA), which integrates DNA microarray and the “refined” ChIP-chip data (derived from applying TRIA to the noisy ChIP-chip data) to reconstruct the transcriptional regulatory modules (TRMs) of the yeast cell cycle process. A TRM is a set of genes that is regulated by a common set of TFs. By organizing the genome into TRMs, a living cell can coordinate the activities of many genes that are needed for a cellular process and carry out complex functions. Therefore, identifying TRMs is useful for understanding cellular responses to internal and external signals. MOFA is helpful for achieving this purpose. We believe that computational analysis of multiple types of data will be a powerful approach to studying complex biological systems when more and more genomic resources such as genome-wide protein activity data and protein-protein interaction data become available.