A genetic algorithm-based protocol for docking ensembles of small ligands using experimental restraints

Area of Science:

• Computational Biology
• Biophysics
• Molecular Modeling

Background:

• Protein-ligand interactions are crucial in biological processes.
• Nuclear Magnetic Resonance (NMR) spectroscopy provides distance restraints for molecular modeling.
• Docking flexible ligands to protein targets remains computationally challenging.

Purpose of the Study:

• To develop and validate a genetic algorithm (GA) for docking flexible ligands to protein receptors.
• To utilize NMR-derived distance restraints within the GA framework.
• To determine the minimum number of ligand conformations required to satisfy experimental constraints.

Main Methods:

• Representing ligand translations, rotations, and dihedral angles as binary strings for GA evolution.
• Employing genetic operators: cross-over, mutation, migration, and selection.
• Defining a fitness function incorporating distance, dihedral, and van der Waals restraints.

Main Results:

• The GA successfully docked ligands to a model system and the streptavidin-biotin complex.
• Low-energy conformations were achieved using simulated intermolecular distance restraints.
• The method accurately identified ligand positions in two distinct binding sites simultaneously.
• No energetic benefit was observed with additional ligand conformations beyond those satisfying restraints.

Conclusions:

• The GA-based method is effective for docking flexible ligands using NMR constraints.
• The approach efficiently determines the minimal ligand orientations for experimental data.
• This method provides valuable structural information for multiple binding sites.