Many quantitative dynamical models in computational systems biology are represented in the form of ordinary differential equations. Most of them are based on known first principles and are highly parametrized. Suitable wet-lab measurements and existing hypotheses about the modelled phenomena can be represented in a temporal logic and considered as constraints for the admissible parameters. Model checking and monitoring techniques known from formal verification can be used to synthesise model parameters that guarantee a given set of temporal properties. In this talk we present the scalable coloured model checking approach for parameter synthesis from LTL and CTL specifications. The method will be presented in the context of formal biochemical space that attempts to avoid the gap between formal methods and biology. On the practical side, application of the method to several biological problems will be given.