Simulating Protein Structures, Complexes, And Dynamics

Computer Research And Technology

Investigators

Linked publications & trials

Abstract

We have continued to develop, implement, and apply simulation methods in computational studies of the energetics and dynamics of biomolecular systems. A study of salt effects on side-chain interactions is in press (J. Phys. Chem. B). We also model proteins based on homology and work to improve the generation and refinement of such models. An invited book chapter discussing homology-modeling issues should appear in the coming months. A paper discussing the prediction of side-chain conformations on protein surfaces is also in preparation. The challenges faced in modeling protein-ligand interactions were discussed in a book chapter (Humana Press, 2005). In collaboration with NICHD we built a model of RNase H1 interacting with an RNA-DNA hybrid (Nucl. Acids Res., 2005). In collaboration with NCI we are studying the interactions involved in kinase binding to Hsp90 (Nat. Struct. Mol. Biol., 2005, as well as a manuscript in preparation). Severeral other projects that involve the use of homology models to guide and interpret experiments are underway. We continue to work on the design of potent HSP90 inhibitors and have begun studying ligands for opioid receptors. In collaboration with NIMH colleagues, quantum chemical calculations were done in search of more potent and selective positron emission tomography (PET) ligands for receptors in the brain (manuscript in preparation). In collaboration with NIDDK, we are characterizing the chemistry, thermodynamics, and spectroscopic properties of carbon nanotubes and their interactions with water. A manuscript is in preparation. Our Langevin simulations of poly-proline peptides in an implicit solvent were essential for the quantitative interpretation of FRET measuments performed by NIDDK colleagues (Proc. Natl. Acad. Sci. USA, 2005). The simulations showed the peptides to be more flexible than expected. Taking the simulated dynamics into account led to a much better description of the measured FRET efficiencies than could be obtained with a simple rigid-rod model of poly-proline. Molecular modeling studies on the A-beta amyloid probes are being carried out to establish a structure-activity relationship. Quantum chemical calculations are also being carried out to quantify the interaction energy between propanediol and water molecules.