hydrophobic interactions between the ligand and the receptor are defined, and
models are generated. Atoms belonging to nonacidic –OH groups (all –OHs
excluding carboxylic, sulphinic, sulphonic, phosphonic or phosphinic acids), –SH
groups, –CC– hydrogens and –NHs (barring trifluoromethyl sulphonamide
hydrogens and tetrazoles) are recognized as HB donor atoms. When such an atom is
found in the distance range of 2.5–3.8 Å from the heavy atom of a HB acceptor of
the receptor molecule, a donor feature consisting of a donor point on the ligand side
and a projected point on the macromolecule side is created. Atoms like –OH
oxygen, –SH sulphur, –CC– carbon or –CN nitrogen are recognized as acceptor
atoms, and an acceptor feature is placed with the initial point positioned on the
acceptor atom and the projected point placed onto the heavy atom of the HB donor
on the receptor within the distance range of 2.5–3.8 Å. The electrostatic interaction
is represented as a vector resembling the definition of the H-bond acceptor.
Hydrophobic areas are implemented in the form of spheres with a tolerance radius
of 1.5 Å located in the centre of hydrophobic atom chains, branches or groups after
testing a group of adjacent atoms to attain a sufficient overall hydrophobicity score.
5.3.2 e-pharmacophore Model Generation by Phase
The e-pharmacophores method of Phase module [46, 88] of Schrodinger suite is a
new approach that utilizes the grid-based ligand docking with energetics (Glide)
extra precision (XP) scoring function [89] to precisely quantify protein–ligand
interactions. XP scoring function calculates enthalpic contribution of each interacting (pharmacophoric) site of a molecule towards the total score. Thus, each site
gets a score based on the sum of enthalpic terms (such as HB, electrostatic,
cation–p, p–p, hydrophobic and hydrophobically packed/associated HBs and other
interactions) and is ranked. Then the e-pharmacophore models are generated from
the top scoring features. The user can choose the number and type of features
required to build a model. E-pharmacophores also include excluded volumes representing the regions of space occupied by the receptor where any portion of the
ligand cannot be accommodated. E-pharmacophores have been shown to screen
diverse set of bioactive molecules as compared to conventional structure-based
methods, making it more useful.
5.4 Dynamic Pharmacophore Model Generation
and Multicopy Simulations
The active sites of the drug targets being very flexible, structure-based pharmacophore models derived from a single conformational state of the protein may not
satisfactorily account for all the possible potential drug–target interactions. In this
situation, molecular dynamics simulation has been a very competent method to
38
C. Choudhury and G. Narahari Sastry
models are generated. Atoms belonging to nonacidic –OH groups (all –OHs
excluding carboxylic, sulphinic, sulphonic, phosphonic or phosphinic acids), –SH
groups, –CC– hydrogens and –NHs (barring trifluoromethyl sulphonamide
hydrogens and tetrazoles) are recognized as HB donor atoms. When such an atom is
found in the distance range of 2.5–3.8 Å from the heavy atom of a HB acceptor of
the receptor molecule, a donor feature consisting of a donor point on the ligand side
and a projected point on the macromolecule side is created. Atoms like –OH
oxygen, –SH sulphur, –CC– carbon or –CN nitrogen are recognized as acceptor
atoms, and an acceptor feature is placed with the initial point positioned on the
acceptor atom and the projected point placed onto the heavy atom of the HB donor
on the receptor within the distance range of 2.5–3.8 Å. The electrostatic interaction
is represented as a vector resembling the definition of the H-bond acceptor.
Hydrophobic areas are implemented in the form of spheres with a tolerance radius
of 1.5 Å located in the centre of hydrophobic atom chains, branches or groups after
testing a group of adjacent atoms to attain a sufficient overall hydrophobicity score.
5.3.2 e-pharmacophore Model Generation by Phase
The e-pharmacophores method of Phase module [46, 88] of Schrodinger suite is a
new approach that utilizes the grid-based ligand docking with energetics (Glide)
extra precision (XP) scoring function [89] to precisely quantify protein–ligand
interactions. XP scoring function calculates enthalpic contribution of each interacting (pharmacophoric) site of a molecule towards the total score. Thus, each site
gets a score based on the sum of enthalpic terms (such as HB, electrostatic,
cation–p, p–p, hydrophobic and hydrophobically packed/associated HBs and other
interactions) and is ranked. Then the e-pharmacophore models are generated from
the top scoring features. The user can choose the number and type of features
required to build a model. E-pharmacophores also include excluded volumes representing the regions of space occupied by the receptor where any portion of the
ligand cannot be accommodated. E-pharmacophores have been shown to screen
diverse set of bioactive molecules as compared to conventional structure-based
methods, making it more useful.
5.4 Dynamic Pharmacophore Model Generation
and Multicopy Simulations
The active sites of the drug targets being very flexible, structure-based pharmacophore models derived from a single conformational state of the protein may not
satisfactorily account for all the possible potential drug–target interactions. In this
situation, molecular dynamics simulation has been a very competent method to
38
C. Choudhury and G. Narahari Sastry