Algorithms and Software for Difficult Proteomics Problems

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The broad, long-term objective of the proposed project is to enable mass spectrometry based protein research. The project will develop algorithms and software for peptide identification in difficult proteomics samples, including heavily modified samples, in which most peptides carry one or more modifications, and mutated samples, in which many peptides differ by one or more amino acid substitutions from the corresponding database peptides. One especially important type of heavily modified sample is a deliberately oxidized sample, used in the technique called "oxidative footprinting" to obtain structural information for proteins and complexes. On difficult proteomics samples, the current peptide identification programs, such as Mascot and SEQUEST, generally give weak results, and there is very little means to assess the quality of the results. Thus the specific aims of the project are: (1) to develop statistical techniques to measure false discovery rates for modification and mutation identifications; (2) to build identification software for oxidative footprinting; and (3) to speed up and improve mutation and modification searching. If the project achieves its aims, biochemistry collaborators will use oxidative footprinting to study antibody-antigen binding for pathogenic bacteria, and will perform deeper and more thorough proteomic analyses of highly variable organisms such as Trypanosoma cruzi. More generally, proteomics laboratories worldwide, working on a wide variety of health related projects, will be able to analyze difficult samples. Researchers will be able to obtain structural information on proteins and complexes that are not amenable to x-ray crystallography or NMR; they will be able to study poorly sequenced and highly variable organisms; and they will be able to doublecheck proteomics analyses by searching for unanticipated chemical modifications. PUBIC HEALTH RELEVANCE: The importance of the proposed project to public health is that it will extend proteomics identifications to more difficult biological samples, such as samples containing unsequenced or poorly sequenced pathogens. The proposed work includes computational tools for "oxidative footprinting", a powerful new technique for studying protein binding and conformations. This technique will enable the study of antibody-antigen binding for pathogenic bacteria; variation in antigens is currently one of the major obstacles to the development of vaccines.