Current Computational Methods for Protein-peptide Complex Structure Prediction.

Area of Science:

• Computational biology
• Drug discovery
• Structural bioinformatics

Background:

• Peptide-mediated protein-protein interactions (PPIs) are vital in biological processes.
• Peptide-based drugs offer high specificity and low toxicity for modulating PPIs.
• Accurate prediction of protein-peptide interactions is essential for drug development.

Purpose of the Study:

• To review recently developed computational methods for protein-peptide docking.
• To present benchmarking datasets and evaluation metrics for docking performance.
• To discuss advanced techniques like molecular dynamics and deep learning for complex prediction.

Main Methods:

• Classification of protein-peptide docking methods into template-based, template-free, and hybrid approaches.
• Compilation of available benchmarking sets for performance assessment.
• Discussion of molecular dynamics simulations and deep learning for complex prediction.

Main Results:

• Categorization of current protein-peptide docking strategies.
• Identification of key datasets and metrics for evaluating docking accuracy.
• Exploration of emerging computational techniques for enhanced prediction.

Conclusions:

• Computational docking methods offer a cost-effective alternative to experimental approaches for studying protein-peptide interactions.
• Standardized benchmarking and evaluation are crucial for assessing and improving docking tools.
• Integration of molecular dynamics and deep learning holds promise for more accurate protein-peptide complex predictions.