Functional organization of the transcriptome in human brain
Michael Oldham
(UC Los Angeles)
Microarrays have emerged as a powerful tool for exploring the functional identities of tissues by
enabling comparisons at the level of the transcriptome. In the brain, transcriptional profiling is
complicated by significant cellular heterogeneity, which can cloud the interpretation of microarray
data and limit the understanding of functional context. New analytical methods that treat
microarray data as a holistic system instead of a collection of discrete measurements have shown
great promise in illuminating the higher-order structure of biological networks. Here we apply one
such method, weighted gene coexpression network analysis (WGCNA), to microarray data derived from
human cerebral cortex, caudate nucleus, and cerebellum. We provide an integrated view of the
transcriptome in each brain region through detailed exploration of gene coexpression relationships.
We demonstrate that the network structure of gene coexpression in human cerebral cortex is highly
reproducible across individuals and microarray platforms. Through comparisons with caudate nucleus
and cerebellum, we identify many aspects of network structure that are conserved across brain
regions, and some that are not. We characterize modules of coexpressed genes that correspond to
each of the major cell classes of the brain: neurons, oligodendrocytes, astrocytes, and microglia.
Other modules distinguish additional cell types, organelles, synaptic function, and gender
differences. Our analysis provides a new foundation for neurogenetic inquiries and reveals the
existence of a previously unrecognized functional organization to the human brain transcriptome.