CAPRI targets T29-T42: proving ground for new docking procedures.

Area of Science:

• Computational Biology
• Structural Biology
• Bioinformatics

Background:

• The Critical Assessment of PRotein Interactions (CAPRI) experiment is crucial for evaluating protein docking algorithms.
• Recent CAPRI targets (T29-T42) included diverse structures (bound, unbound, modeled), posing varied prediction challenges.

Purpose of the Study:

• To assess the performance of protein docking procedures on a range of CAPRI targets.
• To evaluate a new post-docking reevaluation method for improving prediction accuracy.

Main Methods:

• Submitted predictions for CAPRI targets T29-T42, encompassing bound, unbound, and modeled structures.
• Employed a recently developed post-docking reevaluation procedure assessing interface propensity and solvation.
• Analyzed enzyme-inhibitor targets using interface distance from the enzyme centroid.
• Investigated docking ensembles of conformers from normal modes analysis.

Main Results:

• Achieved accurate predictions for targets T40, T41, T42; good for T32; acceptable for T29, T34.
• Docking accuracy generally correlated with prediction difficulty, with modeled proteins yielding less accurate results.
• An accurate prediction was achieved for a modeled tetratricopeptide repeat dimer (T42), an exception to the trend.
• The post-scan reevaluation procedure successfully distinguished correct predictions from false models.
• Interface distance from the enzyme centroid ranked high-quality models for enzyme-inhibitor targets.

Conclusions:

• The developed post-docking reevaluation method effectively refines protein docking predictions.
• Ensemble docking and specific scoring metrics can enhance prediction accuracy, particularly for challenging cases.