Deciphering Long-Range Effects of Mutations: An Integrated Approach Using Elastic Network Models and Protein...

Area of Science:

• Biochemistry and Structural Biology
• Computational Biology
• Genetics

Background:

• Understanding genetic variant impact on protein structure and function is crucial for disease mechanism elucidation.
• The MAVISp framework provides a systematic method for evaluating protein structural effects, including long-range impacts.
• Current methods require refinement for accurate prediction of allosteric variant effects.

Purpose of the Study:

• To critically evaluate and refine the LONG_RANGE module of the MAVISp framework.
• To optimize parameters for detecting significant response sites affected by genetic variants.
• To establish a robust workflow for identifying allosteric protein variants and understanding structural communication.

Main Methods:

• Leveraged data from over 400 proteins to optimize MAVISp parameters.
• Implemented a filtering workflow integrating allosteric free energy, distance constraints, solvent accessibility, and pocket localization.
• Benchmarked results against experimental data from deep mutational scans.

Main Results:

• A 5.5 Å distance threshold was identified as optimal for minimizing local contacts while preserving long-range effects.
• Proposed three metrics to assess protein globularity, addressing limitations of elastic network models for non-globular proteins.
• Demonstrated the potential of molecular dynamics simulations and path analysis for confirming allosteric communication pathways.

Conclusions:

• Established a robust and scalable workflow for detecting allosteric protein variants.
• The refined MAVISp framework enhances the understanding of structural communication in proteins.
• Provides insights into disease-associated protein variants and their functional impact.