Iterative Molecular Dynamics-Rosetta Protein Structure Refinement Protocol to Improve Model Quality.

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• Rosetta is a key tool for high-resolution protein structure refinement.
• Rosetta's effectiveness is limited by conformational sampling, especially for large proteins, potentially trapping algorithms in local energy minima.
• Cryo-electron microscopy (cryoEM) provides low-resolution density maps for large protein complexes.

Purpose of the Study:

• To test if iterating Rosetta with an orthogonal sampling and scoring strategy can improve conformational exploration.
• To develop and present a combined molecular dynamics (MD)-Rosetta protein structure refinement protocol.
• To overcome sampling limitations in protein structure refinement for large proteins.

Main Methods:

• Short molecular dynamics (MD) simulations were performed on models generated by de novo folding of large proteins into cryoEM density maps.
• An iterative refinement protocol combining MD and Rosetta was developed.
• The protocol was tested on four benchmark proteins.

Main Results:

• The combined MD-Rosetta protocol demonstrated an ability to overcome some sampling limitations.
• Two out of four benchmark proteins showed incremental improvement across all three rounds of the iterative refinement.
• MD simulations excelled at subtle rearrangements within secondary structures, complementing Rosetta's focus on side chains and loops.

Conclusions:

• Iterative refinement using MD simulations and Rosetta offers a complementary approach to enhance protein structure modeling.
• This combined protocol facilitates broader conformational sampling, providing an escape route from local energy minima.
• The synergy between MD and Rosetta refines protein structures more effectively, particularly for larger and more complex systems.