Adenoviral (Ad) and adeno-associated viral (AAV) vectors have great promise as gene delivery vehicles for gene therapy and genetic immunization. However, these vectors can non-specifically target tissues and cell types in vivo. Redirected targeting of these vectors by the addition of cell-specific ligands would improve the therapeutic efficacy and safety of these vectors by reducing the effective dosage needed for gene therapy. Phage display technology has been exploited to discover novel cell-specific ligands for vector targeting. However, these ligands are selected in the context of phage and translation of the ligands back into the viral capsid can ablate viral assembly and function or inactivate the targeting function of the ligand itself. To circumvent this ligand-vector compatibility problem, a novel approach to identify cell-specific ligands is described. We have introduced structural "context" onto filamentous bacteriophage and generated random peptide libraries within these contexts for use in ligand selection. The HI loop of the adenoviral capsid was displayed on phage and a random peptide library was generated within this scaffold and used to identify cell-specific ligands against mouse skeletal muscle in vitro. A cell-specific peptide ligand, designated 12.51, was incorporated back into Ad capsid and the redirected Ad vector improved targeting in vitro, suggesting the viability of this approach for ligand discovery. This "context"-based approach was extended towards generating random peptide libraries within streptavidin protein for ligand selection. In addition, a system for conjugation of targeting ligands to the AAV capsid based on the streptavidin-biotin interaction, has been developed. A biotin acceptor peptide was engineered into the AAV capsid and resulted in the development of vectors that are metabolically biotinylated during production in cell lines. This avidin-biotin technology was previously utilized for construction of metabolically biotinylated Ad vectors. However, Ad vectors are extremely immunogenic compared to AAV and may not be suitable for in vivo applications. We constructed metabolically biotinylated AAV vectors and demonstrated proof-of-principle targeting in vitro using various biotinylated ligands. Eventually, streptavidin-context ligands can be conjugated to biotinylated vectors for targeted delivery.