Toward a Holistic View of Transcriptional Regulation

Abstract

The first response to the flood of transcriptomic information obtained from microarray technology was reductionist, identifying individual transcripts that are regulated during an experimental perturbation. This approach continues to identify many previously unsuspected actors in physiological and pathophysiological processes. Coordinate regulation, illustrated by previously ubiquitous heat maps and self-organizing maps, provided some further summary evaluation and suggested associations between genes that had not been previously recognized. Additional connections between regulated transcripts were obtained by mapping them onto known functional pathways, which began to provide mechanistic understanding directly from transcriptional regulation. Over the past few years, a further evaluation of the network of interacting gene regulation has emerged that can, in at least some circumstances, provide a further level of refinement of our understanding of the molecular changes that contribute to disease.1 Article see p 26 One application of network analysis is the statistical evaluation of large amounts of genomic data to derive new insights into previously unsuspected aspects of molecular coordination. The study of Dewey et al2 in this issue of Circulation: Cardiovascular Genetics performs such an analysis on cardiac transcriptional data from a large number of mouse experiments. These investigators use both meta-analysis, the statistical combination of multiple studies, with network analysis, which in this case is applied to the relations of …