ISMARA: Completely automated inference of gene regulatory networks from high-throughput data

As the costs of high-throughput measurement technologies continue to fall, experimental approaches in biomedicine are increasingly data intensive and the advent of big data is justifiably seen as holding the promise to transform medicine. However, as data volumes mount, researchers increasingly realize that extracting concrete, reliable, and actionable biological predictions from high-throughput data can be very challenging. Our laboratory has pioneered a number of methods for inferring key gene regulatory interactions from high-throughput data. For example, we developed motif activity response analysis (MARA)[, which models genome-wide gene expression (RNA-Seq, or microarray) and chromatin state (ChIP-Seq) data in terms of comprehensive predictions of regulatory sites for hundreds of mammalian regulators (TFs and micro-RNAs). Using these models, MARA identifies the key regulators driving gene expression and chromatin state changes, the activities of these regulators across the input samples, their target genes, and the sites on the genome through which these regulators act. We recently completely automated MARA in an integrated web-server (ismara.unibas.ch) that allows researchers to analyze their own data by simply uploading RNA-Seq or ChIP-Seq datasets, and provides results in an integrated web interface as well as in downloadable flat form.