Structure refinement of protein low resolution models using the GNEIMO constrained dynamics method.

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein structure prediction is crucial for understanding function.
• Homology modeling often yields low-resolution models requiring refinement.
• Unconstrained molecular dynamics (MD) has limitations in conformational searching for refinement.

Purpose of the Study:

• To develop and test a constrained MD method for refining low-resolution protein models.
• To evaluate the efficacy of the generalized Newton-Euler inverse mass operator (GNEIMO) algorithm for this purpose.
• To assess the GNEIMO method's performance without experimental constraints.

Main Methods:

• Developed a constrained MD method utilizing the GNEIMO algorithm.
• Modeled proteins as rigid body clusters connected by flexible torsional hinges.
• Employed a "freeze and thaw" clustering scheme within the GNEIMO framework.
• Utilized GNEIMO replica exchange for robust protocol development.

Main Results:

• Demonstrated significant improvement in protein structure refinement from low-resolution decoys.
• Achieved an average ~2 Å root-mean-square deviation (rmsd) improvement for eight test proteins.
• Observed enhanced population density of native-like conformations in GNEIMO trajectories.
• Validated the "freeze and thaw" clustering for localized conformational search.

Conclusions:

• The GNEIMO-based constrained MD method effectively refines low-resolution protein models.
• The method shows promise for enhancing protein structure prediction accuracy.
• The developed protocol is robust and extensible for high-throughput applications.