Computational Drug Discovery and Design

empirical scoring function (SCORE 2.0) [80] and bioavailability by

a set of chemical rules with parameters: molecular weight, logP,

H-bond donor, H-bond acceptor, 160–500,
0.4–5.6, 2–10, and

2–10 was performed.

3.1.4 ADMET Prediction,
Scaffold Hopping,
Molecular Docking,
and Simulation Studies

FAF-Drugs2, an open source web server appliance was used for

investigating the ADMET properties of the designed molecules

[81](seeNote 4). All pharmacokinetic properties were set up

within the acceptable range of Lipinski’s rule of five. Scaffold

hopping was executed using “Chemically advanced template search

(CATS)” to identify matching compounds [82], where two known

VEGFR-2 tyrosine kinase inhibitors Sorafenib and Axitinib were

used as template structures. Further, the newly designed molecules

along with known candidate Sorafenib were taken into the molec-

ular docking platform with the receptor VEGFR-2 tyrosine kinase

and the binding energy was calculated using AutoDock Vina 1.1.2

[83](seeNote 5). Finally, the top ranked protein–ligand complexes

based on the lowest binding energy were simulated for 50 ns to

explore the dynamic behavior and its interatomic interactions that

facilitate the complex stability using Gromacs 4.5.5 software [84].

Thus, top four chemical entities were provided which were novel

compounds toward the target VEGRF-2 tyrosine kinase for anti-

cancer drug design (seeNote 6–8).

4 Notes

1. The nature of search algorithm from a variety of de novo design
   software could provide different kinds of molecules with
   detailed knowledge of chemical properties. It is possible to
   get an optimized compound with improved activity and func-
   tionality with a different framework than the existing.


2. Several software are available which are free, while others are
   commercial [85–88]. These could predict the active sites or
   druggable pockets within the target (protein).

Table 2
(continued)

S. No.

De novo design

software or

program

Year of

publication Concept of function

Receptor

(R)/Ligand

(L) based Reference

47 DOGS 2010 Ligand-based scoring with structural

and pharmacophoric features

L[31]

48 iScreen 2011 Based on molecular docking R [77]

80 Venkatesan Suryanarayanan et al.