There is a need for new therapeutic targets with which to prevent Alzheimer's disease (AD), a major contributor to aging-related cognitive decline. Here we report the construction and validation of a molecular network of the aging human frontal cortex. Using RNA sequence data from 478 individuals, we first build a molecular network using modules of coexpressed genes and then relate these modules to AD and its neuropathologic and cognitive endophenotypes. We confirm these associations in two independent AD datasets. We also illustrate the use of the network in prioritizing amyloid- and cognition-associated genes for in vitro validation in human neurons and astrocytes. These analyses based on unique cohorts enable us to resolve the role of distinct cortical modules that have a direct effect on the accumulation of AD pathology from those that have a direct effect on cognitive decline, exemplifying a network approach to complex diseases.

Suz12 is a component of the Polycomb group complexes 2, 3, and 4 (PRC 2/3/4). These complexes are critical for proper embryonic development, but very few target genes have been identified in either mouse or human cells. Using a variety of ChIP-chip approaches, we have identified a large set of Suz12 target genes in five different human and mouse cell lines. Interestingly, we found that Suz12 target promoters are cell type specific, with transcription factors and homeobox proteins predominating in embryonal cells and glycoproteins and immunoglobulin-related proteins predominating in adult tumors. We have also characterized the localization of other components of the PRC complex with Suz12 and investigated the overall relationship between Suz12 binding and markers of active versus inactive chromatin, using both promoter arrays and custom tiling arrays. Surprisingly, we find that the PRC complexes can be localized to discrete binding sites or spread through large regions of the mouse and human genomes. Finally, we have shown that some Suz12 target genes are bound by OCT4 in embryonal cells and suggest that OCT4 maintains stem cell self-renewal, in part, by recruiting PRC complexes to certain genes that promote differentiation.

Methylation of DNA in combination with histone modifications establishes an epigenetic code that ensures the proper control of gene expression. Although DNA methyltransferases have been shown to interact with histone methyltransferases such as EZH2 (which methylates histone H3 on lysine 27) and G9a (which methylates histone H3 on lysine 9), the relationship between DNA methylation and repressive histone marks has not been fully studied. In cancer cells, promoters of genes are often aberrantly methylated. Accordingly, 5-azacytidine (a DNA demethylating drug) is used for treating patients with myelodysplastic syndrome. However, no genome-scale studies of the effects of this drug have been reported. In this work, we report the effects of 5-azacytidine on global gene expression and analyze ~24,000 human promoters using ChIP-chip to determine how 5-azacytidine treatment effects H3K27me3 and H3K9me3 levels. We found that (1) 5-azacytidine treatment results in large changes in gene regulation with distinct functional categories of genes showing increased (e.g. C2H2 zinc finger transcription factors) and decreased (e.g. genes involved in regulation of mitochondria and oxidoreductase activity) levels; (2) most genes that show altered expression are not regulated by promoters that display DNA methylation prior to the treatment; (3) certain gene classes switch their repression mark upon treatment with 5-azacytidine (from H3K27me3 to H3K9me3 and vice versa); and (4) most changes in gene expression are not due to relief of repression mediated by DNA or histone methylation.

Biological functions are carried out by groups of interacting molecules, cells or tissues, known as communities. Membership in these communities may overlap when biological components are involved in multiple functions. However, traditional clustering methods detect non-overlapping communities. These detected communities may also be unstable and difficult to replicate, because traditional methods are sensitive to noise and parameter settings. These aspects of traditional clustering methods limit our ability to detect biological communities, and therefore our ability to understand biological functions. To address these limitations and detect robust overlapping biological communities, we propose an unorthodox clustering method called SpeakEasy which identifies communities using top-down and bottom-up approaches simultaneously. Specifically, nodes join communities based on their local connections, as well as global information about the network structure. This method can quantify the stability of each community, automatically identify the number of communities, and quickly cluster networks with hundreds of thousands of nodes. SpeakEasy shows top performance on synthetic clustering benchmarks and accurately identifies meaningful biological communities in a range of datasets, including: gene microarrays, protein interactions, sorted cell populations, electrophysiology and fMRI brain imaging.

We compared 12 different cell populations, including embryonic stem cells before and during differentiation into embryoid bodies as well as various types of normal and tumor cells to determine if pluripotent versus differentiated cell types use different mechanisms to establish their transcriptome. We first identified genes that were not expressed in the 12 different cell populations and then determined which of them were regulated by histone methylation, DNA methylation, at the step of productive elongation, or by the inability to establish a preinitiation complex. For these experiments, we performed chromatin immunoprecipitation using antibodies to H3me3K27, H3me3K9, 5-methyl-cytosine, and POLR2A. We found that (1) the percentage of low expressed genes bound by POLR2A, H3me3K27, H3me3K9, or 5-methyl-cytosine is similar in all 12 cell types, regardless of differentiation or neoplastic state; (2) a gene is generally repressed by only one mechanism; and (3) distinct classes of genes are repressed by certain mechanisms. We further characterized two transitioning cell populations, 3T3 cells progressing from G0/G1 into S phase and mES cells differentiating into embryoid bodies. We found that the transient regulation through the cell cycle was achieved predominantly by changes in the recruitment of the general transcriptional machinery or by post-POLR2A recruitment mechanisms. In contrast, changes in chromatin silencing were critical for the permanent changes in gene expression in cells undergoing differentiation.