ATTRACT: protein-protein docking in CAPRI using a reduced protein model

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Accurate prediction of protein-protein complex structures is crucial for understanding biological processes.
• Existing computational methods often struggle with protein flexibility, impacting prediction accuracy.

Purpose of the Study:

• To develop and evaluate a reduced protein model for efficient protein-protein docking.
• To assess the performance of the ATTRACT docking approach in predicting complex structures for CAPRI targets.

Main Methods:

• Utilized a reduced protein model representing amino acids with pseudoatoms.
• Employed the ATTRACT docking approach based on energy minimization.
• Incorporated a multiple conformational copy approach to account for side-chain flexibility.

Main Results:

• Achieved high accuracy (RMSD < 1.8 A) for 3 out of 5 CAPRI targets.
• The reduced model enabled efficient docking of numerous starting structures.
• Identified that both local and global conformational changes are critical for accurate complex prediction.

Conclusions:

• The reduced protein model and ATTRACT approach show promise for protein complex structure prediction.
• Failure to account for global backbone conformational changes can lead to inaccurate predictions.
• Further refinement is needed to fully capture protein dynamics during docking.