Related Experiment Videos
Protein structure elucidation from NMR proton densities.
Alexander Grishaev1, Miguel Llinas
1Department of Chemistry, Carnegie Mellon University, Pittsburgh, PA 15213, USA.
Summary
This study introduces a computational method to determine protein structure from NMR data. It accurately identifies amino acid sequences and side chains, enabling precise structure determination.
Area of Science:
- Structural biology
- Computational chemistry
- Biophysics
Background:
- Nuclear Magnetic Resonance (NMR) spectroscopy generates electron density maps crucial for determining molecular structures.
- Fitting molecular models to these maps is essential for structure elucidation.
- Existing methods face challenges in de novo structure determination from NMR data.
Purpose of the Study:
- To present a novel computational protocol for deriving protein structure from NMR-generated electron density maps.
- To enable accurate identification of polypeptide sequence and atomic positions.
Main Methods:
- Utilizing H(N) atom identification from (1)H/(2)H exchange or HSQC experiments.
- Employing a Bayesian approach to infer sequential connectivity from interatomic distances and chemical shifts.
- Integrating molecular dynamics for structure refinement and selection based on energy criteria.
Main Results:
- Successfully identified polypeptide sequences and side chain atoms, including proline isomers.
- Derived folded protein structures with high accuracy.
- Achieved backbone heavy atom RMSD of 1.0-1.5 Å and all-atom RMSD of 1.5-2.0 Å compared to known structures for collagen 2 and kringle 2 domains.
Conclusions:
- The developed computational protocol provides an effective means for de novo protein structure determination from NMR data.
- The method demonstrates high accuracy and potential for application to various protein systems.
- This approach advances the capabilities of NMR-based structural biology.