Werner Braun, Ph.D., Professor
The challenge in the post genomic era is to make full use of the vast genomic data in understanding the molecular mechanism of the function of proteins. Funding from the Department of Energy has allowed us to develop an expert system, MASIA, to quantitatively identify motifs and patterns of diversity in protein sequences. MASIA, in combination with homology modeling and molecular dynamics simulations, can be used to characterize functional motifs in protein families. In collaboration with Drs. T. Izumi and S. Mitra from the Sealy Center for Molecular Science (SCMS) we are determining how apurinic/apyrimidinic endonucleases respond to subtle differences in the structure and dynamics of DNA sites damaged by low dose ionizing radiation. We are relating motifs identified by MASIA to the function of DNA recognition, catalysis and protein-protein interaction. Screening gene databases with these novel motifs should reveal new candidates for APE related proteins.
In a different avenue we are developing software tools for determining protein structures by NMR in Structural Genomics Projects. Manual analysis of the thousands of cross peaks in protein spectral data is currently the most time consuming part of protein structure determination by NMR. In a project funded primarily by the NSF, we are developing new software tools to automate and integrate all individual steps in the NMR structure determination process. Our NOAH-DIAMOD suite can interpret NOE data in a fraction of the time that would be required for manual assignment, with equivalent or better resolution. Using this method we determined the 3D solution structure of water borne pheromones from the marine mollusk Aplysia by NMR studies in a project with Drs. G. Nagle and S. Painter (Department of Anatomy and Neurosciences). This structure suggests potential receptor-binding sites of these pheromones and we are examining common functional and structural characteristics of pheromones of other species.
The self-correcting distance geometry based tools we developed for NMR have also proven useful for modeling proteins, as demonstrated by our performance CASP4 and CASP5 competition. We are using our methods to model proteins of medical interest in several collaborative projects. We are determining the structural characteristics of cedar pollen allergens in collaboration Drs. E. Brooks, T. Midoro-Horiuti and R. Goldblum (Department of Pediatrics). We have established 3D-database of allergen structures, SDAP, which combines experimental results and our models for other known allergens. In other projects, we have modeled the structure of proteins involved in apoptosis and other forms of cell lysis.