In the last decade computational methods have become one of the most powerful tools in science and technology, offering an unprecedented capability in understanding mechanisms and processes down to the atomic level. One of the areas that have mostly benefited from the swift development of computational capabilities, is Biomedicine. Despite the complexity that usually characterizes the structure and the thermodynamics of biomolecular systems, simulational methods have managed to provide information not easily accessible (or not accessible at all) to pertinent experimental techniques. In particular, as far as it concerns the effort towards designing optimized nanovectors for biomedical purposes, computer simulations have offered the possibility to elucidate the role of different parameters which essentially control the physical properties of such systems. In this presentation we will focus on cases of fully atomistic in-silico studies of polymer and lipid-based constructs for drug and gene-delivery applications, examining structural, dynamic and thermodynamic aspects affecting their physicochemical behavior. The  goal of such studies is to provide new insight towards a bottom-up approach for the design of nanosystems with optimized loading and delivery properties, toward an enhancement of their therapeutical impact.