Contacts-based prediction of binding affinity in protein-protein complexes

Area of Science:

• Biochemistry
• Structural Biology
• Computational Biology

Background:

• Protein-protein interactions (PPIs) are essential for cellular functions.
• Understanding PPI energetics is critical but challenging.
• Current methods lack comprehensive predictive power for binding affinity.

Purpose of the Study:

• To develop a simple yet effective approach for predicting protein-protein binding affinity.
• To investigate the impact of experimental methods and conformational changes on prediction accuracy.
• To enhance prediction by integrating interfacial contact networks with non-interacting surface information.

Main Methods:

• Developed a predictive model based on functional and structural features, specifically the network of interfacial contacts.
• Assessed model performance against a protein-protein binding affinity benchmark.
• Integrated contact-based features with data on the non-interacting surface of proteins.

Main Results:

• The prediction accuracy is significantly influenced by experimental affinity measurement techniques and protein conformational changes.
• A refined model combining interfacial contacts and non-interacting surface data achieved high prediction accuracy on a diverse dataset.
• The proposed method outperformed other tested approaches in predicting binding affinity.

Conclusions:

• The network of interfacial contacts is a key determinant of protein-protein binding affinity.
• Accurate prediction requires careful consideration of experimental conditions and protein dynamics.
• This new approach offers a promising tool for advancing the study of protein-protein interactions.