Docking Technique

Pharmacophore Mapping and

Molecular Modeling Tools

1
 In-Silico Based Design for Drug Discovery

3D structure of the Drug target

Yes No

In-Silico conception 2D QSAR, 3D QSAR,

based on the structure Pharmacophore . . . . .

Structure Based Drug Design Ligand Based Drug Design

(SBDD) (LBDD)
3D target structure is known 3D target structure is unknown
 Molecular docking  Similarity searching
 De novo design  Pharmacophore Mapping
 Molecular dynamics  QSAR
 Protein Data Bank (PDB)
Protein Data Bank (PDB) is the public database where three-
dimensional structures of proteins, nucleic acids, and complex
molecules have been deposited since 1971.

In the last 10 years, the number of 3D structures of the PDB increased

from 48,169 at the end of 2008 to 147,604 in the end of 2018, an
increase of nearly 207%. This implies that in the last 10 years, almost
9,943 new structures have been added to the PDB every year, just
over 27 structures per day, on average.

At the beginning of this decade approximately 25 new entries were

added per day on average. In 2018, over 31 new structures were added
per day, an average daily growth of 24% compared to 2010.

Methods in Computer-Aided Drug Design 8, 343 (2020). 3

Molecular Docking

4
 Protein Ligand Docking

A computational Method which mimics the binding of a ligand to a

protein.
Definition: A docking is a molecular modeling technique that is used
to predict how a protein (enzyme) interacts with small molecules
(ligands).
Two types of docking methods can be applied
1) Rigid docking
2) Semi-rigid docking
5
 Scoring Functions
Scoring function is a key component of a docking tool that guides the search
process and calculates the final binding score.

A scoring function is usually trained by feeding with a collection of compounds

with known bioactivities to capture rules that govern the ligand-target interactions,
so that it is able to predict the binding affinities of any given new interaction.

According to the theories used in scoring function development, the methods can
be divided into four categories:

1) Force field-based scoring functions The energy terms are usually noncovalent
terms such as van der Waals and electrostatic terms.
Example DOCK and AutoDock are based on AMBER force field (classical force
field).

6
 Scoring Functions
2) Empirical-based scoring functions calculate the binding score by summing
up the contributions and penalties of a group of empirically important factors,
such as hydrogen bonding, rotatable bonds, lipophilic contacts, etc, such as
Glide Score, GOLD and X-Score.
3) Knowledge-based scoring functions estimate the binding scores by
summarizing pairwise statistical potentials, which are derived from an inverse
Boltzmann analysis between ligand and its target structure. Scoring functions
of this type include Drug-Score, IT-Score, etc.
4) Machine learning-based, try to encode the features of the protein-ligand
interactions into certain mathematical vectors and train the model by using
sophisticated machine learning techniques, such as neural network (NN),
support vector machines (SVM), and random forest.

7
 Pharmacophore
Mapping/Pharmacophore Modeling

8
 Pharmacophore Mapping
Pharmacophore
A molecular framework that carries (phoros) the essential features
responsible for a drug’s (pharmacon) biological activity.

PABA and p-aminobenzenesulfonamide Analogy between estradiol and

show similar critical distances trans-diethylstilbestrol

An early pharmacophore model for sulphonamides and estradiol

9
Virtual Screening Process Using Pharmacophore Model

An early pharmacophore model for dopamine D1

receptor agonists was developed using ALADDIN

A-68930, a ligand highly selective agonist for the D1

receptor
 Pharmacophore Model of D1 (Dopamine) Agonist

D1 agonist
A-68930

SKF38393 (a dopamine agonist) and dopamine; a crude D1 pharmacophore; the lead discovered
at Abbott; and A-68930, the clinical candidate optimized from that lead.

11
 Pharmacophore Model of Fibrinogen Antagonist
Objective:
The biological hypothesis was based on the observation that the final step in platelet aggregation is
the binding of fibrinogen to GP IIb/IIIa. The key epitope in fibrinogen interacting with that
receptor is a Rat Genome Database (RGD) sequence arginine-glycine-aspartic acid (Arg-Gly-Asp).
A series of peptidic compounds based on RGD, some cyclic-constrained peptides,
had been reported as fibrinogen antagonists.

The Merck’s fibrinogen antagonist

pharmacophore model with a positive charge
separated from a negative charge by 10 to 20 Å
is among the crudest pharmacophores ever
published.

12
 Pharmacophore Model of Protein Kinase C Agonist
Objective:
1) The biological hypothesis was that modulators of PKC, due to PKC’s role in cellular
proliferation pathways, would be useful therapeutics in cancer.
2) The three features were chosen based on known SAR around the known natural-product
ligand, phorbol dibutyrate (PDBu).

The protein kinase C pharmacophore of Wang

et al. and some active hits.

13
The Estrogen Receptor Pharmacophore Model

The estrogen receptor pharmacophore of

Mestres and Veeneman and some latent hits
and their conversion into actual hits.

14
 Pharmacophore Based Virtual Screening of CB1 Antagonist

Objective:
No CB1 crystal structure, only very limited success with homology models.
Aim was to assay 420 compounds selected using a pharmacophore model
Eight CB1 selective antagonists were selected from the
literature including rimonabant .
Pharmacophores were generated with Catalyst. The model
that yielded the most reasonable mapping for Rimonabant
was selected for the database search
The database contained about 500k compounds generated
with Catalyst)

Ki of less than 100 nM.

 Widely Used Docking Tools

Sr. Software Availability Website

No

1 AUTODOCK Free http://autodock.scripps.edu

2 DOCK Free http://dock.compbio.ucsf.edu

3 FlexX Commercial http://www.biosolveit.de

4 GLIDE Commercial http://www.schrodinger.com

5 GOLD Commercial http://www.ccdc.cam.ac.uk

6 iGEMDOCK Free http://gemdock.life.nctu.edu.tw
7 ARGUSLAB Free http://www.arguslab.com
8 FRED Commercial http://www.eyesopen.com

16
 Widely Used Pharmacophore Tools
Sr. No Software Availability
1 PHARMER Free
2 PHARAO Free
3 ALIGN-IT Free
4 CATALYST Commercial
5 MOE Commercial
6 PHASE Commercial

Widely Used Shape-Based Tools

Sr. No Software Availability
1 ROCS and EON Commercial
2 PHASE Commercial
3 SHAPEIT Free
17
 Representative Small-molecule Libraries For Virtual Screening

Database Availability Compounds Website

ZINC Public 980 million http://zinc.docking.org
eMolecules Commercial 7 million http://www.emolecules.com
ChemSpider Public 67 million http://www.chemspider.com
Pubchem Public 111 million http://pubchem.ncbi.nlm.nih.gov
DrugBank Public 13,699 http://www.drugbank.ca
ChEMBL Public 2 million http://www.ebi.ac.uk/chembldb
ChemBridge Commercial 1.3 million http://www.chembridge.com
Specs Commercial 350,000 http://www.specs.net
Asinex Commercial 17,5851 http://www.asinex.com
Enamine Commercial 1.36 billion http://www.enammine.net
Maybridge Commercial 56,000 http://www.maybridge.com
ChemDiv Commercial 1.6 million http://www.chemdiv.com
MCULE Commercial 43 million http://mcule.com/database

18
 Recent Success Stories Making Use of SBVLS Approaches

Drug Target Disease Structure Size of the Initial Computer Tools

(Targeted Function Compound
Mechanism) Collection(s)
without ADME/Tox
Filtering
Neurokinin-1 Mediate Model 827,000 molecules Selector, Unity, and FlexX-
receptor -GPCR numerous Pharm
physiological
processes
Chemokine CCR5 Favorable for Model 1,600,000 molecules Pharmacophore,
receptor agonists protection Surflex and Gold
- GPCR against HIV-1
infection
Potassium ion Wide variety Model 200,000 molecules DOCK, MD
channels of physiological simulation and
processes MM/GBSA
Epidermal growth Cancer X-ray, 315,102 ICM
factor receptor 1M17 molecules
(catalytic site)
Protein kinase Cancer X-ray, 2000 MOE-dock, Glide, Fred
CK2 (catalytic 1JWH natural and Gold
site) compounds

19
 Recent Success Stories Making Use of SBVLS Approaches

Drug Target Disease Function Structure Size of the Initial Computer Tools
(Targeted Compound
Mechanism) Collection(s)
without ADME/Tox
Filtering
Low molecular weight Cancer and X-ray, 875,866 molecules Autodock
protein tyrosine diabetes 1DG9
phosphatase
(catalytic site)
Chymotrypsin-like severe acute Model 361,413 molecules EUDOC
cysteine proteinase respiratory
(catalytic site) syndrome

Chymotrypsin-like severe acute Model 400,000 molecules DOCK,

cysteine proteinase respiratory LigBuilder,
(catalytic site) syndrome Xscore

20
 Recent Success Stories Making Use of SBVLS Approaches

Drug Target Disease Structure Size of the Initial Computer Tools

(Targeted Function Compound
Mechanism) Collection(s)
without ADME/Tox
Filtering
Acetyl-CoA tuberculosis over 4 million DOCK, ICM and the
carboxylase X-ray, molecules ChemDB similarity search
(Catalytic site) 2A7S algorithm
Enoyl–acyl carrier malaria X-ray, 336,600 molecules ICM
protein reductase 1VRW
(Catalytic Site)

Dipeptidyl type 2 X-ray, 800,000 molecules GLIDE

Peptidase IV diabetes 1N1M,
(catalytic site) 1PGQ,
1NU6,
1NU8

21
Examples of Bioactive Molecules Design Using Pharmacophore-
Based Virtual Screening

22
Examples of Bioactive Molecules Design Using Pharmacophore-Based
Virtual Screening

23
Examples of Bioactive Molecules Design Using Docking-Based Virtual
Screening

24
Examples of Bioactive Molecules Design Using Docking-Based Virtual
Screening

25
Examples of Bioactive Molecules Design Using Docking-Based Virtual
Screening

26
Examples of Bioactive Molecules Design Using Docking-Based Virtual
Screening

27