Chemistry Reference
In-Depth Information
VIRTUAL SCREENING
Virtual screening (VS) involves rapid assessment of large libraries of chemical
compounds in order to guide the selection of lead candidates (using computer-
based techniques developed for drug discovery). There is an increasingly larger
chemical space (expanded to more than millions of purchasable, bioactive
compounds). VS yields (low false hit rates)/(good yields) such that synthesis and
testing tasks are manageable [418-546].
The most extensively used VS methods are docking (1), pharmacophore (2),
quantitative structure activity relationship (QSAR) (3), similarity searching (4)
and machine learning (5). These methods can be grouped as follows: a) structure-
based virtual screening (SBVS);
i.e.
docking of candidate ligands into a protein
target followed by applying a scoring function to estimate the likelihood that the
ligand will bind to the protein with high affinity; b) ligand-based virtual screening
(LBVS) which applies computational descriptors of molecular structure,
properties, or pharmacophore features to analyze relationships between active
templates and compounds from chemical libraries [324, 418-456].
In the first method, (1), active compounds can be identified geometrically by
docking to a pre-selected target site followed by binding configuration
optimization and scoring (enabling the identification of novel active compounds
with their corresponding binding modes). Comprehensive conformational
sampling, adequate modeling of solvation and entropic effects as well as
improvement of scoring functions can be used to address the methods limitation in
modeling target structural flexibility.
The classical docking-based VS approach begins with the usage of a three-
dimensional structure of the target protein. Subsequently, usage of docking
software as well as compounds can be stored in databases (docked inside the
target protein). Scoring functions can be used to analyze the docking results. The
ligands that interact better inside the receptor are selected and the most
representative subjected to binding assays or
in vitro
inhibition. Beyond the
classical docking-based VS approach, there is also usage of docking software as
the central core of more complicated approaches.
