Our research comprises three interrelated lines of investigation:

• A recurring theme in systems biology research is the integration of large amounts of high-throughput data, for example from gene expression studies (Sontrop et al, PLoS ONE, 2011). For this purpose we develop software for the construction of disease-specific compendia of high-throughput data.

• Next to the integration of high-throughput measurements, the collection of prior biological knowledge, e.g., published information about biochemical pathways is another key component in systems biology research. We recently performed a systematic comparison of five comprehensive and often used metabolic networks and uncovered surprisingly large differences between these databases (Stobbe et al, BMC Systems Biology, 2011). We are also working on the integration of our findings in an international effort to build a consensus human metabolic network (Thiele et al, Nature Biotechnology, 2013).

• The previous two research lines provide the building blocks for the development of computational methods to uncover molecular networks perturbed by disease. Disease-specific networks are constructed using our compendium database and curated pathway databases. We focus on the development of network-based methods for the analysis of gene expression data across (a) different tissues and (b) different organisms. An example of the former is our collaboration with the Verhoeven group (Medical Biochemistry) on the interorgan coordination in response to fasting and nutritional overload (Hakvoort et al, Journal of Biological Chemistry, 2011). An example of the latter is our NBIC project for investigating similarities and differences between model organisms and human using omics data and disease-specific functional networks. Our goal is to provide a systematic framework that will be useful for understanding the applicability of rodent models for human metabolic syndrome.