Restrained LES Method

The binding of a ligand to a receptor plays a crucial role in most biological processes, such as signal transduction, viral cell entry, transport of various species, DNA repair, as well as in the development of drugs for the treatment of diseases. It is therefore important to simulate the binding poses and affinities between ligands and the receptors, especially if this can be done more quickly or efficiently than by experiment. For example, a longstanding problem in the design of new pharmaceuticals is computational screening of a very large number of trial drug molecules before attempting more costly

experimental investigations.  As the cost of MD simulations decreases due to advances in computer speed, ligand screening methods that rely on finite-temperature MD simulations (rather than rigid-body or flexible docking or minimization) are becoming more practical in drug design.  One popular ligand screening method involves simulating multiple copies of (possibly chemically distinct) ligands within a single copy of the target protein, so that multiple ligand binding sites can be predicted for the ligands simultaneously.  The method is known as Locally Enhanced Sampling (LES) or, alternatively, Multiple Copy Simultaneous Search (MCSS).  Despite the apparent simplicity of LES, the binding site predictions do not rigorously correspond to free energy minima, essentially because the target protein is influenced by all of the ligands in a mean-field sense.  We have shown how the LES method can be extended to provide a rigorous link to computing free energy minima, which, in the context of protein-ligand interactions, would correspond to finding the most stable binding sites on the protein.  The present limitation of the approach is that it can be rigorously applied to multiple copies of chemically identical ligands only.

Reference :

V. Ovchinnikov and M. Karplus. A restrained locally enhanced sampling method (RLES) for finding free energy minima in complex systems. (preprint)