Advances in nanotechnology have the potential to utilize biological and polymeric systems to address fundamental scientific and societal issues, including molecular electronics and sensors, energy-relevant light harvesting, â??greenâ?? catalysis, and environmental cleanup. In many cases, synthesis and fabrication are well within grasp, but designing such systems requires simultaneous consideration of large numbers of degrees of freedom including structure, sequence, and functional properties. In the case of protein design, even simply considering amino acid identity scales exponentially with the protein length. This work utilizes computational techniques to develop a fundamental, molecularly detailed chemical and physical understanding to investigate and design such nano-assemblies. Throughout, we leverage a probabilistic computational design approach to guide the identification of protein sequences that fold to predetermined structures with targeted function. The statistical methodology is encapsulated in a computational design platform, recently reconstructed with improvements in speed and versatility, to estimate site-specific probabilities of residues through the optimization of an effective sequence free energy. This provides an information-rich perspective on the space of possible sequences which is able to harness the incorporation of new constraints that fit design objectives. The approach is applied to the design and modeling of protein systems incorporating non-biological cofactors, namely (i) an aggregation prone peptide assembly to bind uranyl and (ii) a protein construct to encapsulate a zinc porphyrin derivative with unique photo-physical properties. Additionally, molecular dynamics simulations are used to investigate purely synthetic assemblies of (iii) highly charged semiconducting polymers that wrap and disperse carbon nanotubes. Free energy calculations are used to explore the factors that lead to observed polymer-SWNT super-structures, elucidating well-defined helical structures; for chiral derivatives, the simulations corroborate a preference for helical handedness observed in TEM and AFM data. The techniques detailed herein, demonstrate how advances in computational chemistry allot greater control and specificity in the engineering of novel nano-materials and offer the potential to greatly advance applications of these systems.