Author: Gunnar Cedersund, Fraunhofer-Chalmers Research Centre.
Names and affiliation of other authors:
Mats Jirstrand; Fraunhofer-Chalmers Research Centre, Gothenburg, Sweden
Jacob Roll; Department of Electrical Engineering, Linköping University, Sweden
Peter Strålfors; Department of Cell Biology, Linköping University, Sweden
Oral or poster: Poster
The first part of this talk presents a new modelling framework denoted core-box modelling. As the name indicates the result is a combination of a minimal core model with a mechanistic grey-box model. The advantages of the core model is that it is characterised how well-established its different predictions are with respect to the available knowledge (data and mechanistic characterisations). Further, the core model is often obtained using hypothesis testing which allows for non-trivial conclusions during the development phase. The advantage of the grey-box model is that it contains all known mechanisms in the system. This is useful, e.g., when using the model for detailed drug-testing scenarios. Finally, a translation between the two models is established; this allows them to be considered as two aspects of the same model, and for their individual strengths to be combined.
The second part of this talk shows the application of this framework to the development of a core-box model for insulin signalling in human fat cells. The hypothesis testing during the development of the minimal model shows
that the recycling of the insulin receptor into the cytosol, and the dephosphorylation and eventual recycling to the membrane, are all necessary to display the observed behaviour. The core-box model shows some robust predictions, e.g. regarding the speed of the internalisation. Finally, the advantages of the core-box model are demonstrated on examples for in silico drug-testing.