We can sequence any DNA of interest, including full genome of an individual, but we are not making full use of this information yet. The ability to predict phenotype occurring for given genetic background and environmental conditions will revolutionise medicine and biotechnology. I am convinced that Molecular Biology knowledge will be used to reverse engineer molecular machinery of the cell as a computer model and use mechanistic simulation to predict cellular behaviour for particular set of genetic and environmental perturbations. I will present examples of detailed quantitative models of well studied network modules in bacterial cells and discuss current limitations of quantitative dynamic simulations at whole-cell scale. Subsequently, I will present constraint-based modelling of genome scale metabolic networks with SurreyFBA software. Finally, I will introduce Quasi Steady State Petri Net (QSSPN) – a new hybrid simulation algorithm allowing multi-formalism simulation integrating i) qualitative rule based ii) stochastic kinetic iii) deterministic kinetic and iv) constraint based models. I will present dynamic simulation of molecular interaction network describing gene regulation, signalling and whole-cell metabolism in human hepatocyte.