Chemistry Reference
In-Depth Information
experimental testing. HTVS studies can however be challenging if there is need to
properly prepare million compounds databases. Currently, many of such curated
databases have been adequately designed and are available for research.
Each chemical compound in the library used in HTVS is at, in principle, a
possible pose. The compounds can be sequentially docked at the protein (Target).
The molecules docked with the lowest free energies of interactions are ranked in
the list and considered as potential 'hits' to be possible subjects to further
experimental assays. Virtual Screening programs typically consists of algorithms
that search the space available to candidate molecules and a function representing
a measure of binding affinity. The minimization process yields solutions for active
molecules in accordance with experiment and separate active/inactive compounds.
When we speak of docking, we typically refer in general to a computational
technique which places in the receptor (macromolecular target) a small ligand
(molecule) and determines the binding affinity as well. The main components of a
receptor-ligand structure
in silico
are a) docking (orientation and conformational
sampling constrained by the binding site of the receptor) b) scoring,
i.e.
for a
molecule the best ligand orientation, conformation and translation (pose). If a
ligand database is docked, scoring functions are used to rank the best order of
ligands. A successful docking should accurately predict the binding affinity as
well as the pose.
Before starting on project, it is necessary to assess the docking, virtual screening
setup,
via
literature or actually performing it to determine if the choice of
program, scoring functions, and techniques for receptor and ligand flexibility are
validated. There may be difficulties in positioning low-resolution structures.
There may be an absence of alternative ligand structures (protonation, tautomeric
states, and orientation and conformation compatibility with electron density).
Crystal structures are important and docking quality can be strongly targeted and
scoring dependent.
We can speed up finding good potential candidate drugs by using prediction
methods based on theoretical computing methods and molecular modeling to
determine the three-dimensional structure for new hit leads. Prediction accuracy
