Photon energy absorption and electronic energy transfer (EET) represents the first fundamental step in both natural and artificial light-harvesting systems. The most striking example is photosynthesis, in which plants, algae and bacteria are able to transfer the absorbed light to the reaction centers in proteins with almost 100% quantum efficiency. Recent two-dimensional spectroscopic measurements suggest that the role of the environment (a protein or a given embedding supramolecular architecture) is fundamental in determining both the dynamics and the efficiency of the process. What is still missing in order to fully understand and characterize EET is a new theoretical and computational approach which can reproduce the microscopic dynamics of the process based on an accurate description of the playing actors, i.e. the transferring pigments and the environment. Such an approach is a formidable challenge due to the large network of interactions which couples all the parts and makes the dynamics of the process a complex competition of random fluctuations and coherences. Only a strategy based upon an integration of computational models with different length and time scales can achieve the required completeness of the description. This project aims at achieving such an integration by developing completely new theoretical and computational tools based on the merging of quantum mechanical methods, polarizable force fields and dielectric continuum models. Such a strategy in which the fundamental effects of polarization between the pigments and the environment will be accounted for in a dynamically coupled way will allow to simulate the full dynamic process of light harvesting and energy transfer in complex multichromophoric supramolecular systems.