S4MPLE—Sampler for Multiple Protein-Ligand Entities: Methodology and Rigid-Site Docking Benchmarking

Molecules

Laurent Hoffer¹  Camelia Chira²  Gilles Marcou¹  Alexandre Varnek¹  Dragos Horvath¹ 

[1] Laboratoire de Chemoinformatique; UMR 7141, Université de Strasbourg, 1 rue B. Pascal, Strasbourg 67000, France; E-Mails:;Department of Computer Science, Technical University, Cluj-Napoca 400027, Romania; E-Mail:

关键词: genetic algorithms;    conformational sampling;    docking;    interaction fingerprints;    force field fitting;   

DOI  :  10.3390/molecules20058997

来源: mdpi

【 摘 要 】

This paper describes the development of the unified conformational sampling and docking tool called Sampler for Multiple Protein-Ligand Entities (S4MPLE). The main novelty in S4MPLE is the unified dealing with intra- and intermolecular degrees of freedom (DoF). While classically programs are either designed for folding or docking, S4MPLE transcends this artificial specialization. It supports folding, docking of a flexible ligand into a flexible site and simultaneous docking of several ligands. The trick behind it is the formal assimilation of inter-molecular to intra-molecular DoF associated to putative inter-molecular contact axes. This is implemented within the genetic operators powering a Lamarckian Genetic Algorithm (GA). Further novelty includes differentiable interaction fingerprints to control population diversity, and fitting a simple continuum solvent model and favorable contact bonus terms to the AMBER/GAFF force field. Novel applications—docking of fragment-like compounds, simultaneous docking of multiple ligands, including free crystallographic waters—were published elsewhere. This paper discusses: (a) methodology, (b) set-up of the force field energy functions and (c) their validation in classical redocking tests. More than 80% success in redocking was achieved (RMSD of top-ranked pose < 2.0 Å).

【 授权许可】

CC BY   
© 2015 by the authors; licensee MDPI, Basel, Switzerland.