Accelerated Structural Prediction of Flexible Protein-Ligand Complexes: The SLICE Method.

Area of Science:

• Computational Chemistry
• Structural Biology
• Biophysics

Background:

• Predicting protein-ligand binding is crucial for drug discovery.
• Flexible protein targets pose significant challenges for traditional structural prediction methods.
• Existing molecular dynamics approaches can be computationally expensive and time-consuming.

Purpose of the Study:

• To present a novel computational method, SLICE (Selective Ligand-Induced Conformational Ensemble), for predicting the induced-fit binding of ligands to flexible protein targets.
• To model ligand-induced structural adaptations in protein binding sites using a combination of docking and molecular dynamics.
• To validate the SLICE method on known protein-ligand systems and challenging flexible targets.

Main Methods:

• SLICE combines opportunistic stochastic jumps of ligand position with standard molecular dynamics.
• Conformational jumps are selected from structures generated by docking software (AutoDock Vina).
• Molecular dynamics (Amber code) is used to relax the protein structure and generate new poses.

Main Results:

• SLICE successfully modeled induced-fit binding configurations for tested systems, including CBX8-H3K9Me3.
• The method showed promising results compared to long classical and accelerated dynamics approaches.
• Further optimization of SLICE parameters is needed for precise crystal structure replication in some cases.

Conclusions:

• SLICE offers a computationally efficient and promising approach for predicting induced-fit binding events, particularly for flexible protein targets.
• The method provides a valuable tool for structural prediction in computational chemistry and drug discovery.
• Ongoing research will focus on optimizing SLICE parameters for enhanced accuracy and broader applicability.