Use of surface area computations to describe atom-atom interactions

Area of Science:

• Structural biology
• Biophysics
• Computational biology

Background:

• Accessible surface area (ASA) and atomic contact area (ACA) are key metrics in protein structure analysis.
• Current applications primarily use ASA to quantify the hydrophobic effect's contribution to protein stability.
• Extending ACA's utility requires establishing a direct link to protein stability and interaction energies.

Purpose of the Study:

• To investigate the relationship between atomic contact areas and the free energy of atom-atom interactions.
• To determine if atomic contact area changes correlate with experimental protein stability data.
• To develop a predictive model for protein stability based on atomic interactions.

Main Methods:

• Utilized a linear model linking protein stability to buried accessible surface area between atom pairs.
• Assessed the model's validity using 124 lysozyme mutants with known X-ray structures and experimental stability.
• Tested residue representations with varying numbers of atom types (4-9 types).

Main Results:

• Simple residue representations (4-5 atom types) showed no clear linear relationship between stability and buried accessible area.
• A clear linear relationship was observed for representations with 6-9 atom types, minimizing heterogeneity.
• A simplified linear model using a basic function for buried accessible area was explored for potential use in protein structure prediction.

Conclusions:

• The number of atom types used in residue representation significantly impacts the correlation between atomic contact areas and protein stability.
• A refined representation with 6-9 atom types establishes a robust linear relationship, enhancing the predictive power of atomic contact area analysis.
• The findings suggest that atomic contact area analysis, particularly with optimized representations, can be a valuable tool for understanding and predicting protein stability and interactions.