Computer Studies of Protein Structure and Function

Columbia University Health Sciences, New York NY

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): The goals of the proposed research include: a) The development of methods and associated computational tools that use three-dimensional structural information to predict protein function. b) The development of methods for the integration of structural information in the generation of networks of protein- protein interactions (interactomes). c) The application of structural information in the creation of an interactome of human and tumor B-cells and the use of this network in applications to B-cell Biology. These research goals are motivated by a number of factors. First, there is a general sense that structural information, as provided for example by the Protein Structure Initiative (PSI) is not fully exploited by the wider biological community. An interactive function prediction server is being developed with this issue in mind. Second, Structural Biology has not been fully integrated into Systems Biology approaches to predict protein networks and it would be extremely valuable to try to increase the integration of the two fields. Third, human B-cell phenomena are of great biological and medical interest and there is an exciting opportunity to incorporate structural information into this field in a novel way. A central element of the approach to be taken is the use of structural alignments to reveal novel functional relationships between proteins that have been classified as belonging to different protein folds. Evidence is provided that there is a wealth of functional information in such remote relationships and computational tools to mine and visualize this information will be developed. The research design also includes the description of a novel approach to use structural alignments to identify potential protein-protein ligand, protein-protein and protein-DNA binding partners and to identify interfacial residues in the complexes they form. A novel scoring scheme for protein-protein interactions will be developed and a database of potential complexes, designed originally for human proteins, will be constructed. The database will also identify complexes that are unlikely to form. The information in this database will be incorporated into a Baysian inference scheme and used in the construction of a structure-enabled B-cell Interactome. PUBLIC HEALTH RELEVANCE: The relevance of the proposed research to public health lies in part in the insights, and accompanying methods, that will be provided about structure/function relationships in proteins. Specific applications to B-cell Biology have direct relevance to understanding a variety of B-cell related cancers.