Predicting the Effect of Mutations on Protein-Protein Binding Interactions through Structure-Based Interface...

Area of Science:

• Biochemistry and Structural Biology
• Computational Biology and Bioinformatics

Background:

• Protein-protein interactions are vital for cellular functions.
• Mutations disrupting these interactions can lead to severe cellular consequences.
• Experimental determination of protein binding affinity is challenging for large-scale studies.

Purpose of the Study:

• To develop and validate a computational method for predicting the impact of mutations on protein-protein binding affinity.
• To assess the efficacy of a scoring function based on interface structure profiles.

Main Methods:

• Utilized a scoring function based on interface structure profiles from analogous protein-protein interactions in the Protein Data Bank (PDB).
• Developed a composite model by integrating sequence-derived, residue-level coarse-grained potentials with the interface structure profile score using random forest training.
• Compared the accuracy of the interface profile score and the composite model against established methods.

Main Results:

• The interface profile score alone demonstrated accuracy comparable to all-atom potentials but was significantly faster.
• A composite model combining interface profile scores with other potentials achieved a Pearson correlation coefficient >0.8 for predicted versus observed binding free-energy changes.
• The developed method does not necessitate high-resolution atomic models of mutant structures.

Conclusions:

• Interface structure profiling is a powerful and efficient tool for predicting mutation effects on protein binding affinity.
• The composite model offers high accuracy, comparable to or exceeding current state-of-the-art methods.
• This approach has significant potential applications in recognizing disease-associated mutations and in protein interface design.