Specific protein-protein interactions are required for virtually all cellular functions. Nonetheless, the molecular basis for specificity in these interactions remains incompletely understood. The proposed research will explore factors leading to specificity in coiled-coil interactions. Coiled coils provide a useful model system for the study of specific interactions among proteins and for the development of new biochemical methods for characterizing these interactions. In addition, as coiled-coil domains are found in a wide variety of proteins of biological and medicinal interest, the detailed study of this class of proteins has implications for our understanding of phenomena ranging from transcription to cytoskeletal structure and function. We propose a method for using sequence-specific DNA binding as a probe for helix orientation preference and for partner specificity in coiled coils. This strategy allows us to generate and quantitatively test large numbers of mutant coiled coils efficiently. It also permits us to take advantage of in vivo selection techniques. Specifically, we describe an approach for selecting antiparallel coiled coils from randomized protein libraries. No other method is currently available for distinguishing between parallel and antiparallel coiled coils in vivo. The rules we identify for helix orientation preference and partner specificity will lead to an improved understanding of protein molecular recognition and facilitate the design of coiled-coil proteins with improved or novel recognition properties. In addition, these studies will also be crucial for translating the primary sequence of putative coiled-coil proteins into structural and functional information.