UC San Diego

As the availability of protein crystallographic structures grows, new methods are needed to characterize, compare, and interpret crystal structure data. Surface geometry is important in molecular function, as interactions with other molecules in the cell happen at the protein surface. This dissertation describes a method of mapping and comparing surfaces of crystallographic structures, using the protein kinase family as a model. Pockets are rapidly computed using two pieces of software, FADE and Crevasse. FADE uses gradients of atomic density to locate grooves and pockets on the molecular surface. Crevasse, a new piece of software, segments the FADE output into distinct pockets that can be used in computations. A map of pockets on the catalytic core of Protein Kinase A shows the ATP and peptide docking sites, locations for C-helix anchoring, the myristylation site, and sites for regulatory subunit interaction. There are other sites identified where Protein Kinase A is likely to interact with other proteins, but the binding partners have not been identified. Spatially clustering pockets across a family of aligned proteins can identify similarities and differences within the family. In the set of ten kinase cores studied, differences in the active site cleft between serine/ threonine and tyrosine kinases are visible. Shared mechanisms of C-helix anchoring are also evident, and a novel site at the top of the N-lobe is present in all the kinases. There are other pockets on the kinases that are strongly conserved but have not yet been mapped to a protein-protein interaction