Virtual Screening by Christoph Sotriffer

Author:Christoph Sotriffer
Language: eng
Format: epub, mobi
Publisher: John Wiley & Sons
Published: 2011-03-27T16:00:00+00:00

SCARE [76] systematically replaces pairs of neighboring residues by alanines intending to remove the area of the site that hinders the ligand to intrude. It thus generates an ensemble consisting of multiple gapped pocket versions. The ligand is docked to each pocket version and the best scoring pocket is directed to a refinement stage in which the backbone with the original side chains is optimized and rescored.

8.3 Flexible Protein Handling in Docking-Based Virtual Screening

The aim of protein–ligand docking is to predict the pose that a ligand will adopt upon binding. Moreover, docking would like to estimate the binding free energy. In contrast to that, the docking engine that accomplishes a virtual screening has to fulfill a slight different task. The prediction of the optimal complex geometry is less in focus. In fact, docking each compound from a library, it has to decide which of them will probably bind and which will not. This requires a scoring function that ranks the compounds, optimally, according to their binding free energy. Furthermore, the docking engine has to evaluate a library sometimes containing hundreds of thousands of compounds and therefore it should screen rapidly to ensure a high-throughput of the process.

Fully flexible docking methods are a step toward more reliable compound assessment. For their application in structure-based virtual screening, some effects have to be considered that result from the introduction of protein flexibility. The following sections retrospectively consider the above-mentioned fully flexible docking approaches in respect to their efficiency and ability to discriminate binders from nonbinders.

8.3.1 Efficiency of Fully Flexible Docking Approaches in Retrospective

To make library evaluation affordable, a fast docking engine is an absolute premise. In general, modeling target flexibility is counter-productive for the acceleration of the docking calculation. For computational methods, the time limiting step is the modeling of protein conformational change. However, with respect to VS, the increase of computational costs mainly depends on in which docking stage the flexibility is introduced – prior, during, or posterior to the actual docking run. An introduction of target flexibility as a preprocessing step is rather affordable than postprocessing. In the former case, the expensive conformational change has to be modeled only once and the prepared target structure can be reused in later VS experiments. In the latter case, the whole range of complexed library compounds should be optimized individually for an objective ligand evaluation. This implicates that docking approaches following conformational selection should be more efficient for the employment in VS. If the ensemble is already generated, the runtime scales at most linearly with ensemble size that may be further reduced by using grids, united protein structures, or sophisticated search algorithms.

The practicability of an induced fit docking technique in VS mainly depends on the degree of flexibility of the considered target structure. The docking stage in consecutive induced fit approaches, that is, the soft docking stage, is quite efficient. The costly step is that of complex optimization afterward. Nevertheless, approaches optimizing only a restricted set of side chain conformations may be affordable too, either in consecutive or simultaneous fully flexible docking.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.