This project is a collaborative effort with scientists in two disciplines, computer science and biology. Such a collaboration is essential for the success of the development of computational tools and their later use in biological research. The first crucial step, which is to understanding the biological problem at hand in mathematical terms and to translate this into a mathematical problem with a clear objective, provides the foundation of this proposal. In this project, the resulting mathematical problems are tackled by the computer scientist and their solutions are implemented and then used in biological research. These computational tools are of interest in their own right and are likely to find application outside the biological area. The focus of this work is a study of alpha-helical coiled coil proteins. The streptococcal M-protein serves as a model protein. For the majority of proteins, the secondary structure of the protein, which is critical for its proper function, cannot simply be resolved by examining the sequence of the amino acids that define the protein. While x-ray crystallography is the only reliable means available today by which to solve the structure of a protein molecule, it cannot always be used. The alpha-helical structure is commonly found in proteins because of its ability to stabilize proteins through short regions of helix-helix packing, referred to as a coiled-coil interaction. The main goal of this project is to develop computational tools that enable the comparison of surface characteristics of two M proteins, and more generally of any two coiled-coil proteins. Algorithms are devised to identify identical amino acid clusters on the outer surface of coiled-coil proteins. Such clusters, which are not readily identified through linear sequence homology, could predict the formulation of cross-reactive epitopes between microbial and mammalian molecules . This is then verified by the synthesis of the peptide region and a variety of serological tests. Algorithms are also designed to identify and display tandem repeat blocks in sequences. At the DNA level, these represent mutational `hot spots.` In bacterial proteins the function of these repeats is not fully understood, but perhaps they have evolved as a means of mutating orders of magnitude more frequently than spontaneous mutation events. Finally, the sequence of M proteins from strains of the same serotype, (M6), which were isolated over the last century are compared.
