Protein docking by the interface structure similarity: how much structure is needed?

Area of Science:

• Computational Biology
• Structural Bioinformatics
• Biophysics

Background:

• Co-crystallized protein-protein complexes enable template-based modeling for protein-protein docking.
• Structure alignment is key for identifying remote similarities between protein targets and templates.

Purpose of the Study:

• To systematically determine the optimal interface definition and size for structure alignment-based protein-protein docking.
• To establish guidelines for effective protein docking applications, especially for large-scale analyses.

Main Methods:

• Conducted a large-scale study evaluating various interface cutoff values for structure alignment.
• Utilized a benchmark set of 99 protein complexes to assess docking success rates.
• Analyzed the impact of interface size on the representation of structural details.

Main Results:

• Interface areas with cutoff values less than 12 Å inadequately represent critical structural details.
• Increasing the cutoff beyond 12 Å did not significantly improve success rates for high-accuracy models and decreased them for lower-accuracy models.
• A 12 Å cutoff was identified as optimal for interface alignment-based docking.

Conclusions:

• The 12 Å cutoff provides an optimal balance for capturing protein interface structural information in docking.
• This finding offers practical guidelines for protein-protein docking approaches, including high-throughput and genome-scale applications.
• The study highlights the importance of interface definition for successful computational modeling of protein interactions.