Integrating Cross-Linking Experiments with Ab Initio Protein-Protein Docking.

Area of Science:

• Structural biology
• Computational biology
• Biochemistry

Background:

• Ab initio protein-protein docking aims to predict complex structures but often requires experimental validation.
• Experimental data, such as chemical cross-linking followed by mass spectrometry (CX-MS), can provide valuable constraints for docking.

Purpose of the Study:

• To integrate CX-MS experimental data into ab initio protein-protein docking algorithms.
• To enhance the accuracy and success rate of predicting protein complex structures.

Main Methods:

• Developed an approach combining protein-protein docking with CX-MS data as distance constraints (Euclidean or void-volume).
• Tested the method on 19 protein complexes from the Protein Data Bank with identified cross-links.
• Implemented strategies for assigning cross-links to specific interfaces in multi-component complexes and utilized symmetry information.

Main Results:

• The rank of top-scoring near-native predictions improved at least twofold compared to docking without cross-link data.
• The success rate for top 5 predictions nearly tripled.
• The approach demonstrated an effective balance between retaining correct predictions and reducing false positives.

Conclusions:

• Integrating CX-MS data into protein-protein docking is a powerful strategy for improving complex structure prediction.
• The developed methods for handling multi-component complexes and symmetry enhance prediction performance.
• This integrated approach offers a more reliable way to identify protein complex structures.