Current State-of-the-art for Virtual Screening and Docking Methods - New Developments in Medicinal Chemistry - page 98

MedusaDock is a flexible docking approach which models both ligand and
receptor flexibility simultaneously with sets of discrete rotamers. The program has
developed an algorithm which builds the ligand rotamer library 'on-the-fly'
during docking simulations demonstrating a rapid sampling efficiency and high
prediction accuracy in both self- (to the cocrystallized state) and cross-docking (to
a state cocrystallized with a different ligand). The latter mimics the virtual
screening procedure in computational drug discovery [518].
LigDockCSA is a protein-ligand docking software which uses a powerful global
optimization technique, conformational space annealing (CSA) and a scoring
function that combines the AutoDock energy and the piecewise linear potential
torsion energy [519].
bhDock is a blind hierarchical docking method which uses two-step algorithms. It
starts with a comprehensive set of low-resolution binding sites determined by
analyzing entire protein surface and ranking by a simple score function.
Subsequently, ligand position is determined via a molecular dynamics-based
method of global optimization starting from a small set of high ranked low-
resolution binding sites. Refinement of the ligand binding pose starts from
uniformly distributed multiple initial ligand orientations and uses simulated
annealing molecular dynamics coupled with guided force-field deformation of
protein-ligand interactions to determine the global minimum [520].
NEURODOCK does automated docking of flexible molecules into receptor
binding sites by ligand self-organization in situ . It was developed for the analysis
of potential binding poses of structurally complex flexible ligands [521].
ICM methodology proceeds with internal coordinates in order to optimize, in a
grid-based receptor field, flexible ligands. The potentials of the grid includes
hydrogen bonds, electrostatic, hydrophobic and van der Waals terms. ECEPP/3
force field and MMFF partial charges are used to calculate energies. Global
optimization begins with random conformational change of angles, free bonds and
torsions according to the biased probability Monte Carlo algorithm. This followed
by local energy minimization of the analytical differentiable terms [522].