Common Hits Approach: Combining Pharmacophore Modeling and Molecular Dynamics Simulations

Abstract

We present a new approach that incorporates flexibility based on extensive MD simulations of protein-ligand complexes into structure-based pharmacophore modeling and virtual screening. The approach uses the multiple coordinate sets saved during the MD simulations and generates for each frame a pharmacophore model. Pharmacophore models with the same pharmacophore features are pooled. In this way the high number of pharmacophore models that results from the MD simulation is reduced to only a few hundred representative pharmacophore models. Virtual screening runs are performed with every representative pharmacophore model; the screening results are combined and rescored to generate a single hit-list. The score for a particular molecule is calculated based on the number of representative pharmacophore models which classified it as active. Hence, the method is called common hits approach (CHA). The steps between the MD simulation and the final hit-list are performed automatically and without user interaction. We test the performance of CHA for virtual screening using screening databases with active and inactive compounds for 40 protein-ligand systems. The results of the CHA are compared to the (i) median screening performance of all representative pharmacophore models of protein-ligand systems, as well as to the virtual screening performance of (ii) a random classifier, (iii) the pharmacophore model derived from the experimental structure in the PDB, and (iv) the representative pharmacophore model appearing most frequently during the MD simulation. For the 34 (out of 40) protein-ligand complexes, for which at least one of the approaches was able to perform better than a random classifier, the highest enrichment was achieved using CHA in 68% of the cases, compared to 12% for the PDB pharmacophore model and 20% for the representative pharmacophore model appearing most frequently. The availabilithy of diverse sets of different pharmacophore models is utilized to analyze some additional questions of interest in 3D pharmacophore-based virtual screening.

Molecular docking and network connections of active compounds from the classical herbal formula Ding Chuan Tang

“…Pharmacophore models are constructed from multiple pharmacological features and targets and are used in computational drug discovery to recognize the molecular features of one or more compounds (ligand) required for a lock and key fit with macromolecules (proteins) with the same or similar biological activity (Thangapandian et al, 2011; Wieder et al, 2017). The DS 4.5 software contains a Ligand pharmacophore profiler (LPP) protocol, allowing for virtual screening of multiple chemical compounds against a pharmacophore database (PharmaDB) containing 117,423 pharmacophore models and constructed from 7,028 protein databank (PDB) protein-ligand X-ray crystal structures.…”

PyRod Enables Rational Homology Model‐based Virtual Screening Against MCHR1

“…Molecular dynamics (MD) simulation can be used to address such artifacts and additionally, provide valuable information about the flexibility and thermodynamic properties of the system . PyRod, a free and open‐source Python software, combines the strength of MD simulations with structure‐based 3D pharmacophore searches by analyzing the protein environment of water molecules in protein binding pockets and subsequently generates pharmacophore features for virtual screening …”

Comparing pharmacophore models derived from crystallography and NMR ensembles

“…A growing trend is to use more than one protein structure when creating pharmacophore models to represent the conformational flexibility inherent to each system. The structures can come from crystallography [1,7,8,10,14], NMR [1], or molecular dynamics (MD) simulations [6,9,11–16]. The resulting pharmacophore models from each structure can be combined into a single, overarching model [7,8,11,12,14,16] or used as a set.…”

“…The resulting pharmacophore models from each structure can be combined into a single, overarching model [7,8,11,12,14,16] or used as a set. When used as a set, scientists can simply combine all hits [15], look for common hits across the pharmacophore models [6], or use machine-learning algorithms to develop a combined model for ligand binding [9,10,13]. …”