In recent years, the analysis of cell signalling systems through data-based models in ordinary differential equations (ODE) or other paradigms (e.g. stochastic models) has emerged as an invaluable tool to understand the underlying complexity of the protein interactions happening in cellular signal transduction. Compared with other biochemical systems, the modelling of cell signalling systems faces additional difficulties related to the challenges quantifying protein-protein processes but also to the lack of complete information about the topology of the considered network interactions. Since in most of the metabolic systems the complete network of interactions is (virtually) perfectly established, in cell signalling systems the real structure of the pathways is an open question to be elucidated either in parallel or through mathematical modelling based analysis. In this talk we discuss the use of power-law models (advantages and challenges) in biochemical systems. We also show how pre-existent biological knowledge and quantitative data can be integrated through mathematical modelling to validate hypothesis about the structure of signalling pathways.