The recent significant progress within molecular biology, genetics and bioinformatics has resulted in detailed knowledge, and measurement data, of the components that constitute a living cell. However, the various functions of a cell can not be attributed to single genes or proteins, but are rather a result of interactions between DNA, RNA, proteins and metabolites within highly complex networks. The network interactions are primarily created through biochemical reactions combined with transport mechanisms, and involve a substantial amount of feedback type interactions. Unravelling the functionality of biochemical networks, and thereby linking cell physiology to the information encoded in the genome, will require a systems approach based on mathematical modelling. In this talk we introduce the use of systems and control theoretical concepts to causal (dynamic) modelling and analysis of intracellular biochemical networks. We briefly present some modelling approaches based on experimental data, possibly combined with physical knowledge. One important use of such models is in analysis aimed at detecting specific subnetworks, and mechanisms, underlying given functions. As an example of such, we present a method we have recently developed for identifying mechanisms underlying multistability (switches) and periodic oscillations in biochemical networks. We illustrate the method by application to some examples, including cell cycle control, circadian rhythms and the oscillatory metabolism of activated neutrophils. Referanser: 1. Modellering: FEBS J., May 2005 http://www.s3.kth.se/publications/2005/1727.pdf 2. Analyse, mest for regulerings- og kjemiteknikere: IEEE CS, Aug 2004 http://www.s3.kth.se/control/main/general/publicationdb/documents/jacobsen0401.pdf 3. Analyse, med litt intro til reguleringsteknikk for bio: Systems Biology, June 2004 http://www.s3.kth.se/control/main/general/publicationdb/documents/jacobsenschmidt0401.pdf