Grant: $148,828 - National Institutes of Health - Aug. 31, 2009
No votes have been cast for this award yet
Award Description: PAR 03-106 Innovations in Biomedical Computational Science and Technology
Project Description: The overall project focuses on 3D modeling of protein-protein interactions, which is one of the most important problems in computational structural biology. The project follows the current major shift of the protein docking paradigm towards knowledge-based techniques. The increased emphasis on the knowledge-based aspect, due to the continuing rapid development of the field and new results generated in our laboratory, is supported by the Supplement funding. In one month since the beginning of the support, we started generating the training and validation sets of protein-protein complexes for the machine learning-based scoring algorithm. The DOCKGROUND database is being used for selection of the complexes according to the criteria that characterize typical protein-protein complexes (interface area, size of the chains, complex type, structure/sequence redundancy, and such). In addition, we have started selection of the representative features of protein structure and sequence for the machine learning procedure that will be capable of distinguishing near-native and false-positive protein-protein matches.
Infrastructure Description: N/A
Jobs Summary: No jobs were created or retained this quarter. (Total jobs reported: 0)
Project Status: Less Than 50% Completed
This award's data was last updated on Aug. 31, 2009. Help expand these official descriptions using the wiki below.