Protein structure refinement via molecular-dynamics simulations: What works and what does not?

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein structure determination is crucial for understanding biological function.
• Accurate protein models are essential for drug discovery and molecular biology research.
• The Critical Assessment of protein Structure Prediction (CASP) competition benchmarks structure prediction methods.

Purpose of the Study:

• To evaluate protein structure refinement strategies employed by the Feig group during CASP11.
• To analyze the factors influencing refinement success, including sampling and initial model quality.
• To develop an improved protocol for generating high-quality protein structures.

Main Methods:

• Utilized molecular dynamics (MD) simulations for protein structure refinement.
• Implemented an enhanced selection and averaging protocol for MD trajectories.
• Assessed structural quality using the MolProbity score and analyzed refinement of secondary structure elements.

Main Results:

• Achieved modest overall refinement, with significant improvements for some protein targets.
• Demonstrated that refinement success correlates with the amount of sampling and initial model quality.
• Developed an updated protocol capable of generating high-quality structures with low MolProbity scores.

Conclusions:

• The developed protocol offers an efficient method for refining protein structures.
• Accurate stereochemical features can be achieved with modest computational resources.
• The study provides insights into optimizing protein structure refinement techniques for future predictions.