WISDOM: A Grid-Enabled Drug Discovery Initiative against Malaria - Grid Computing: Infrastructure, Service, and Applications - page 354

Docking is the method of i rst choice for rapid in silico screening of large
ligand databases for drug research, since it is based on a rational physical
model. Basically, protein-compound docking is about computing the
binding energy of a protein target to a library of potential drugs using a
scoring algorithm. The target is typically a protein that plays a pivotal role
in a pathological process; for example, the biological cycles of a given
pathogen (parasite, virus, bacteria, etc.). The goal is to identify which
molecules could dock on the protein active sites in order to inhibit its
action and therefore interfere with the molecular processes essential for
the pathogen. Libraries of compound 3D structures are made openly
available by chemistry companies that can produce them. Many docking
software are available either open-source or licensed.
However, there is very often a compromise between speed and accuracy
of results (in terms of the actual binding mode as well as the calculated
afi nity values) concerning the best scoring docking solutions. Docking
methods usually generate a number of possible orientations of ligands in
the binding site of the receptor, and the “correct” one (e.g., the orientation
observed in the crystal structure) may not necessarily be ranked among
the i rst docking solutions. This is due to dei ciencies in orientation sam-
pling and to the approximate nature of the scoring functions. Thus, it seems
reasonable to subject screening results to postdocking rei nement; for
example, using molecular dynamics or similar methods that can describe
biomolecular structure and energy in more details.
An example of workl ow to deploy virtual screening that additionally
includes postdocking rei nement goes through the following steps:
Step 1 Selection of the target, the chemical compound database,
and the docking software.
Step 2 Preparation . If the selected target is an X-ray crystal struc-
ture with a bound ligand, then it requires preparing the binding
site of the protein by taking 6-8 Angstroms from the cocrystal-
lized ligand, ensuring that signii cant amino acids for the activity
are included in the binding site. Information on the signii cant
amino acids can be obtained either from the literature or from the
Brookhaven protein database. Regarding the preparation of com-
pounds to dock, open-source databases providing ready to dock
molecules are made available on the Internet by companies sell-
ing the compounds, but both target and compound have to be
prepared according to the needs of the software.
Step 3 Docking . Access to data analysis and visualization software
is required at this point.
Step 4 Postdocking analysis . Results are analyzed based on the
docking energy score and binding mode of the compound inside
the binding site.