As technological advances allow a better identification of cellular networks, more and more molecular data are produced allowing the construction of detailed molecular interaction maps. One strategy to get insights into the dynamical properties of such systems is to derive compact dynamical models from these maps, which would then be handled more efficiently for the analysis of their dynamics.Starting from two specific case studies of biochemical network, I will present a methodology for the derivation of qualitative dynamical models from biochemical networks. Properties are formalised using abstraction interpretation techniques. We first abstract states and traces by quotienting the state space by intervals. The induced abstract semantics is not precise enough to reproduce the properties of interest of our two examples. We then refine the abstract semantics by introducing additional constraints and information on the kinetics computed with an arithmetics derived by abstract interpretation. The semantics we obtained is finally able to reproduce our properties of interest.