Solvents to Fragments to Drugs: MD Applications in Drug Design

Abstract

Molecular dynamics(MD) simulations are playing an increasingly important role in structure-based drug discovery (SBDD). Here we review the use of MD for proteins in aqueous solvation, organic/aqueous mixed solvents (MDmix) and with small ligands, to the classic SBDD problems: binding mode and binding free energy predictions. The simulation of proteins in their condensed state reveals the solvent structure and preferential interaction sites (hot spots) on the protein surface. This information is largely transferable across all classes of protein ligands (from water to drugs) and can be used very effectively to understand ligand recognition and improve the predictive capability of well-established methods such as molecular docking. MD simulations for protein and drug or drug-like compounds are now being used but are still computationally expensive and can only be applied to specific cases. On the other hand, MDmix simulations can now be used in SBDD and we will describe the latest developments and implementations. We expect to see an increase in the application of these techniques to a plethora of protein targets to identify new drug candidates with the advent of new tools and faster computers.

grand: A Python Module for Grand Canonical Water Sampling in OpenMM

“…2 The prevalence of water-mediated interactions between proteins and small molecules was highlighted by a 2007 study, in which it was found that 85 % of 392 high resolution crystal structures analyzed contained at least one water molecule bridging the protein and ligand. 3 Molecular dynamics (MD) is now routinely used in computer-aided drug design, 4 given the level of atomistic detail that can be obtained regarding the dynamics of protein structures.…”

Structure-based molecular modeling in SAR analysis and lead optimization

“… [61] Focuses on the specifics of docking calculations with covalent ligands. [62] Discuss applications and practical aspects of MD simulations with mixed solvents. [63] Provides an overview of protein-peptide docking methods and outlines their capabilities, limitations, and applications in structure-based drug design.…”

Using machine‐learning‐driven approaches to boost hot‐spot's knowledge

“…Kozakov et al showed that only fragments with a good spatial overlap with top‐ranked HS were expected to be extended to larger, useful ligands 89 . However, when dealing with a shallower protein–protein interface, the lack of protein flexibility may introduce demanding problems to the detection of key HS 90 . Molecular dynamics (MD) application, although very useful to overcome this issue, continues to be time consuming, and as such the systematically use of organic/aqueous mixed solvents has been proposed to predict binding modes and affinities, or to guide the fragment evolution process.…”