Combined ligand and structure based design- While primarily a
structure based design program FRED can also use a known bound ligand within
the active site to improve results.
Receptor Site Shape- FRED has a very effective method for
determining the shape of an active site (that works well even on very
shallow/open binding sites), which allows it to very efficiently reject
poses in incorrect positions. This technology can also be used to detect
active sites on a protein when the site is not known a priori.
Consensus Structure- This method of selecting a correctly
docked pose from a set of likely candidates uses the consensus of multiple scoring
functions. The consensus structure method improves the chances that the pose
selected is correct (relative to using any single scoring function). Please note
that "consensus structure" uses multiple scoring functions to compare different
poses of the same ligand, unlike "consensus scoring" which is used for
ligand-ligand comparison.
M.A.S.C. scoring- MASC stands for Multiple Active Site Correction
and is a method of correcting for systematic biases in any scoring function.
The MASC method corrects for systematic bias by comparing a ligands score
to the same ligands score in a set of standard reference sites. Ligands
which score well in the reference sites are assumed to be promiscuous, and
are penalized.
Chemgauss scoring- Chemgauss is a scoring function developed by
OpenEye which uses Gaussian functions to describe the shape and chemistry of
molecules. Chemgauss is currently on version 3 and is described in more detail
in the Scoring function chapter.
Exhaustive Docking- FRED's docking strategy is to exhaustively
score all possible positions of each ligand in the active site. The exhaustive
search is based on rigid rotations and translations of each conformer. This
novel strategy completely avoids the sampling issues associated with stochastic
methods used by many other docking programs.
Most of these technologies are described in more detail in the Theory section.
