On the applicability of elastic network normal modes in small-molecule docking

Area of Science:

• Computational Biology
• Structural Bioinformatics
• Drug Discovery

Background:

• Protein flexibility is crucial for accurate protein-ligand docking.
• Normal Mode Analysis (NMA) effectively models collective protein motions, but its utility in small-molecule binding prediction remains under-explored.
• Existing studies on NMA for protein-ligand interactions are limited.

Purpose of the Study:

• To conduct a large-scale investigation into the applicability of Normal Mode Analysis (NMA) for protein-ligand docking.
• To assess the impact of using NMA-generated conformations on docking performance.
• To determine if NMA can reliably predict conformational changes upon small-molecule binding.

Main Methods:

• Utilized 433 apo/holo protein-ligand pairs from the Astex data set.
• Generated holo-like conformations from apo structures using the first few normal modes.
• Performed docking simulations using AutoDock, GOLD, and FlexX on these generated conformations.
• Evaluated docking performance based on the number of normal modes used.

Main Results:

• NMA-generated conformations, even in a best-case scenario, showed restricted utility for protein-ligand docking.
• Docking performance did not consistently improve with an increasing number of normal modes.
• No straightforward rule emerged for selecting the optimal number of normal modes for effective docking.

Conclusions:

• Normal Mode Analysis (NMA) has limited applicability in enhancing protein-ligand docking accuracy for small molecules.
• Predicting conformational changes upon ligand binding using NMA is challenging and lacks a generalizable strategy.
• Further research is needed to refine methods for incorporating protein flexibility into docking protocols.