The methodology for systematic building of coarse-grained molecular models from detailed atomistic simulations, and, in principle, from ab-initio simulations is discussed. Detailed atomistic simulations are used to extract different kinds of structural information about the system, which is then used to determine effective potentials for coarse-grained model of the same system. The statistical-mechanical equations expressing canonical properties in terms of potential parameters can be inverted and solved numerically according to the iterative Newton scheme.

Thermodynamic compatibility by the obtained in this way effective potentials is discussed. The approach is illustrated on several examples: parametrization of atomistic ion-water potentials from ab-initio simulations; effective implicit solvent potentials between ions and between ions and DNA, effective potential for coarse-grained lipid model, and a few others. Finally, a novel software MagiC implementing computations of effective potentials for coarse grained model of arbitrary structure from atomistic trajectories by the inverse Monte Carlo method is presented.