Use of prior knowledge in biological systems modelling

An enormous amount of biological knowledge has been generated by the scientific community and is available from a large number of biological databases, scientific literature, and domain experts. This knowledge is actively used to define new hypotheses and to validate new findings, but it may also be included in computational modelling and high-throughput data analysis as prior knowledge in order to improve the analysis or guide it towards meaningful solutions. In this thesis we explored the use of prior knowledge in data-driven and knowledge-driven modelling approaches for high-throughput data analysis and biological systems modelling. In the first part of this thesis we reviewed methods that incorporate prior knowledge in statistical models for the analysis of high-throughput transcriptomics and metabolomics data. In the second part we used prior knowledge to model two biological systems. First, we used sparse prior knowledge to build a network-based model of a multi-organ genistein elimination pathway that can assist in the design of new experiments. Secondly, we developed a mathematical model of B-cell affinity maturation that generated valuable insights in the affinity distribution among (un)expanded subclones and allowed to study the development of B-cell lineage trees. In summary, we explored possibilities to facilitate high-throughput data analysis with prior knowledge, and demonstrated the use of prior knowledge in biological systems modelling.