Bridging between NMA and Elastic Network Models: Preserving All-Atom Accuracy in Coarse-Grained Models

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein dynamics offer insights into functional mechanisms.
• All-atom simulations of large protein complexes are computationally expensive.
• Existing coarse-grained models often lack realistic dynamics.

Purpose of the Study:

• To develop an efficient method for constructing coarse-grained models with all-atom accuracy for protein dynamics.
• To address the limitations of current coarse-grained potentials in capturing realistic molecular motions.

Main Methods:

• Utilized the sparseness of the Hessian matrix for computational efficiency.
• Developed a novel iterative matrix projection approach.
• Applied the method to the large GroEL/GroES protein complex.

Main Results:

• Achieved all-atom level accuracy in describing normal mode motions for large biomolecular complexes.
• Demonstrated high computational efficiency through the novel iterative matrix projection method.
• Provided new insights into the GroEL/GroES mechanism, linking conformational transitions to low-frequency modes.

Conclusions:

• The developed method enables accurate and efficient simulation of protein dynamics for large complexes.
• This approach bridges the accuracy gap between all-atom and coarse-grained models.
• The findings offer a powerful tool for understanding the functional mechanisms of large protein machinery.