Formalizing Coarse-Grained Representations of Anisotropic Interactions at Multimeric Protein Interfaces Using Virtual...

Area of Science:

• Computational Biology
• Biophysics
• Molecular Modeling

Background:

• Simulating large biomacromolecular complexes is computationally intensive.
• Traditional coarse-grained models struggle with anisotropic interactions at protein interfaces.
• Existing methods often require higher-order potentials, increasing computational cost.

Purpose of the Study:

• To develop a computationally efficient method for representing directional interactions in coarse-grained models.
• To improve the accuracy of molecular simulations for multimeric protein assembly.
• To enhance the prediction of protein-protein interface interactions.

Main Methods:

• Introduction of virtual sites to represent directional interactions.
• Optimization of virtual site parameters using relative entropy minimization.
• Comparison of virtual site models against traditional coarse-grained models using two protein assembly case studies.

Main Results:

• Virtual site models demonstrate higher fidelity in predicting pairwise correlations.
• The models accurately predict the assembly behavior of multimeric protein complexes.
• Morphologically consistent assembly predictions align with experimental data.

Conclusions:

• Anisotropic interaction representation is crucial for accurate macromolecular assembly simulations.
• Virtual sites offer a computationally efficient approach to model these interactions.
• This work enables more accurate and efficient coarse-grained simulations of biological assemblies.