pyDockRescoring Server

pyDockRescoring Server rescores docked protein-protein poses to improve identification of near-native interaction models for structural biology, disease mechanism analysis, and drug design.

Key Features:

• Rescoring Capability: Rescores docking outputs, including poses from pyDockWEB, using advanced scoring techniques to prioritize near-native conformations.
• IRaPPA Integration: Integrates the IRaPPA (Information Retrieval for Protein-Protein Association) pipeline to recast docking selection as an information retrieval problem.
• Algorithmic Basis: Adopts internet search ranking and electoral voting system methodologies to evaluate and rank docked structures based on calculated biophysical properties.
• Performance Statistics: Demonstrated identification of a near-native structure within the top 10 solutions for up to 50% of benchmarked complexes and within the top 100 for up to 70% across four different docking methods.
• Cross-Platform Implementation: IRaPPA has been implemented for use with other docking programs such as SwarmDock and ZDOCK.

Scientific Applications:

• Protein Function Analysis: Produces more accurate protein-protein interaction models to aid interpretation of protein collaboration in cellular processes.
• Disease Mechanism Exploration: Helps identify how mutations or perturbations disrupt protein-protein interfaces by pinpointing near-native structures.
• Drug Design and Discovery: Provides ranked interaction models useful for designing inhibitors or mimetics that target protein-protein interfaces.

Methodology:

Integrates calculated biophysical properties into a scoring system inspired by information retrieval techniques, using internet search ranking and electoral voting concepts to prioritize docked poses.