Today, biology is less concerned with individual macromolecules than with the way these entities interact. Two important questions arise from the mass of data becoming available on cellular function: (1) How do the thousands of macromolecular components present in the cell manage to interact with the correct partners to form a dynamically stable system capable of transmitting information and reacting to changes inside and outside the cell; (2) How are the molecular machines that underlie most cellular processes assembled and how do they carry out their functions? Despite progress in computer power, it is not possible to study such complicated problems with conventional simulation techniques and new approaches are needed to push towards larger multi-component systems and slower processes. Moving from atomic-scale representations to coarse-grain models is one way to solve this problem. How these models should be developed and parameterized and how one can move between atomic and coarse-grain models are the subjects of this workshop. Progress in this area requires a multi-disciplinary approach with contributions from mathematical, physical, computational and biological scientists. Success will strongly influence the future relevance of molecular simulations for systems biology and the life sciences in general. Our workshop is organized around formal lectures, but leaves ample time for discussions.