Biologically enhanced sampling geometric docking and backbone flexibility treatment with multiconformational...

Area of Science:

• Computational biology
• Structural bioinformatics
• Molecular modeling

Background:

• Predicting protein-protein interactions is crucial for understanding biological processes.
• Existing docking methods often struggle with accuracy and efficiency.
• Incorporating biological and experimental data can enhance docking predictions.

Purpose of the Study:

• To develop and validate an efficient, biologically enhanced sampling geometric docking method.
• To improve the prediction of protein-protein binding modes using the FTDock algorithm.
• To integrate diverse data sources and account for protein backbone flexibility.

Main Methods:

• Utilizing a biologically enhanced sampling geometric docking approach based on FTDock.
• Incorporating active site data from experimental and theoretical analyses into rotation-translation scans.
• Employing a sphere-based discretization of proteins to bias docking solutions towards relevant interfaces.
• Implementing a multiconformational superposition scheme to model protein backbone flexibility.

Main Results:

• The method demonstrated improved performance on CAPRI targets, yielding more hits and lower ligand root-mean-square deviations (RMSDs) during conformational search.
• For Target 19, significant improvements in hit numbers and RMSDs were observed.
• For Target 18, multiconformational superposition increased near-native structures from 1 to 53 and reduced the minimum ligand RMSD from 8.1 Å to 2.9 Å.

Conclusions:

• The biologically enhanced sampling geometric docking method, incorporating protein flexibility, enhances the accuracy of protein-protein binding mode prediction.
• This approach shows promise for improving computational drug discovery and structural biology.
• The method successfully reproduced and improved upon results for challenging CAPRI targets.