COMP 28 |
| The sequence-specific recognition of RNA by proteins plays a fundamental role in gene expression. While many experimental studies have explored the molecular basis for the sequence dependence of protein-RNA recognition, a few studies have also explored this problem from a computational perspective. However, a quantitative description of protein-RNA interactions remains to be developed. We have developed an orientational dependent hydrogen bonding potential based on the occurrence of hydrogen bonding contacts and orientation in the structures of protein-nucleic acid complexes as well as an all-atom distance-dependent statistical potential function, and applied both to the prediction of protein-nucleic-acid interactions from structure. In both protein-DNA and protein-RNA systems, both methods are able to reliably discriminate native complexes from near-native decoys. The statistical potential can infer the relative binding energies of a number of experimentally characterized mutations to protein-RNA complexes as well, and is able to predict the sequence recognized by RNA-binding proteins. The sensitivity to fine structural details makes this statistical potential an inviting choice for problems requiring the detailed computational modeling of protein-DNA and RNA interactions. These include structure-based genome annotation and the rational design of proteins with altered nucleic-acid recognition specificities. We are currently applying this approach to the redesign of the specificity of a set of paradigmatic RNA-binding proteins. |
|
Protein-Nucleic Acid Interactions: Experimental and Modeling Analysis
8:30 AM-12:15 PM, Sunday, August 19, 2007 BCEC -- 156B, Oral
Division of Computers in Chemistry |