Alzheimer's disease (AD) is the most common form of dementia, affecting millions of aging people worldwide with no known cure. Transcriptional studies suggest that AD is a dysfunction of many cell types and multiple cellular pathways, but there is no consensus on which genes in these pathways are disease-relevant or on whether or not these pathways are causal. Furthermore, no attempt has yet been made at combining results from these studies in a way that also accounts for phenotypic information. In this dissertation, three studies are presented that provide insight into AD pathophysiology from a systems biological perspective. In the first study, gene coexpression networks characterizing the progression of AD and normal aging are created. This study finds that dysfunction of many of the same processes occurs with AD progression as with normal aging, only to a much greater extent. For example, groups of coexpressed genes (modules) corresponding to both synaptic and mitochondrial functions show decreased expression with both age and markers for AD progression. In the second study, gene coexpression networks for both mouse and human brain are created, in order to address the issue that many mouse models for AD only partially reproduce the disease pathophysiology. This study finds multiple lines of evidence suggesting links between neurodegenerative disease and glial cell types in human, including human-specific correlation of presenilin-1 with oligodendrocyte markers, and significant enrichment for known neurodegenerative disease genes in microglial modules. In the final study, to address the issue of regional vulnerability in AD, a new transcriptional analysis of both non-demented and AD brain is presented, including both CA1 and the relatively less affected CA3 as a regional control. In addition to confirming results from many previous transcriptional analyses, this study finds a significant linear relationship between region and disease, suggesting that transcription levels in a region may reflect that region's vulnerability to disease and providing candidate neuroprotective and vulnerability genes. Modules for major cell types are also found that each show distinct disease-relevant expression patterns, including one that suggests microglial activation may be an early event in AD progression. By utilizing gene coexpression network methodologies in the study of AD, novel insights into the onset and progression of this disease can be found.