Accepted Manuscript

Title: Docking Techniques in Pharmacology: How Much

Promising?

Authors: Meenakshi Gupta, Ruchika Sharma, Mahakpreet

Singh, Anoop Kumar

PII: S1476-9271(17)30696-5
DOI: https://doi.org/10.1016/j.compbiolchem.2018.06.005
Reference: CBAC 6879

To appear in: Computational Biology and Chemistry

Received date: 30-9-2017

Revised date: 21-2-2018
Accepted date: 30-6-2018

Please cite this article as: Gupta M, Sharma R, Singh M, Kumar A, Docking Techniques
in Pharmacology: How Much Promising?, Computational Biology and Chemistry
(2018), https://doi.org/10.1016/j.compbiolchem.2018.06.005

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 Docking Techniques in Pharmacology: How Much Promising?

Meenakshi Gupta1, Ruchika Sharma2, Mahakpreet Singh3, and Anoop Kumar1*

1
Department of Pharmacology, Indo-Soviet Friendship Pharmacy College (ISFCP),
Moga, Punjab, India.
2
Department of Biotechnology, Indo-Soviet Friendship College of Professional Studies

(ISFCPS), Moga, Punjab, India.

T
3
Department of Pharmacy Practice, Indo-Soviet Friendship Pharmacy College (ISFCP),

IP
Moga, Punjab, India.

R
* Corresponding author: Anoop Kumar, Associate Professor, Department of Pharmacology,
ISFCP, Moga, Punjab -142001, India. Email id: abitmesra@gmail.com.

SC
Graphical Abstract
U
N
A
M
ED
E PT
CC
A

1
Highlights
• Docking studies play an important role in the field of pharmacology.
• It predict the energetically favourable binding conformations of ligands in the active
site cavity of particular receptor
• The accuracy of these studies is questionable if not used properly (selection of correct
binding sites of the target protein, the selection of correct docking pose etc).
• Many compounds show high dock score, but fail in molecular dynamics (MD)

T
simulation which question on the reliability of docking.

IP
Abstract

R
In the drug development process, large numbers of molecules are failing at a later stage due

SC
to safety & efficacy concerns which are generally investigated by conducting various
experiments on animals but unfortunately, animal experimentation is associated with lots of

U
limitations. Thus, there is an on-going interest in the development of alternative methods of
N
pharmacological evaluations that requires less number of animals. Docking is an alternative
approach for screening of compounds before actually testing it on animals. It is the best
A
option to predict the interaction of drug molecules towards particular receptors and has
M

become of high interest to the scientific community to save time and money involved in the
drug development process. The use of these techniques is increasing day by day among
ED

researchers due to availability of user friendly software’s. The number of papers related to
docking has dramatically increased over the past decades. A variety of docking software’s are
currently available to predict the interaction of molecule towards receptors, but there is a lack
PT

of knowledge and confidence among scientists regarding accuracy of these softwares. There
are many issues to be considered when undertaking a docking study such as selection of
E

correct binding sites of the target protein, the selection of correct docking pose, lack of clarity
CC

over whether the compound is an antagonist or agonist. Many compounds show high dock
score, but fail in molecular dynamics (MD) simulation which arise the question on reliability
A

of docking. These problems should be cause for caution and concern before performing
docking. Therefore, to ensure accurate and effective application of docking techniques, it is
necessary to understand the method’s strengths, limitations, the scope of application, and
interpretation. In this review, we try to summarize fundamental concepts, why these
techniques are needed, various docking methods and recent developments in this area.
Further, current challenges and future perspectives of these tools have been also discussed.

2
Keywords: Docking study; Molecular dynamics; Pharmacology.
1. Introduction
Docking study is the application of mathematical, biological and computer based models to
predict the binding affinity of small molecules towards particular receptor [1]. According to
various researchers, docking study is strictly a computer-based analysis which predicts the
binding affinity of new chemical entities (NCE) or drugs on the basis of their chemical
structure. However, docking study is not merely this concept. This research integrates
advances in molecular biology, biotechnology, bioinformatics, mathematics, chemistry with

T
modeling, and computer science in order to increase the predictive power of docking

IP
softwares [2]. As we all know, drug development process is long and expensive as well as

R
attrition rate of the NCE is very high especially at the later stages. Further, the increasing
incidence rate of disease and resistance of microbes to the approved drugs requires speeding

SC
up of the drug development process. To save the time and money of drug discovery and
development process, docking studies are widely used to predict the interaction of designed

U
analogues toward particular receptor [3]. Recent reports have indicated the potential of these
N
studies in the field of pharmacology. In pharmacology, these methods could predict the
A
affinity of drugs within the binding site of the target of interest. These methods could also
predict the pharmacokinetic property of drug on the basis of Lipinski rule of five and
M

interactions of the drug towards particular metabolic enzyme which can provide a better
understanding about complexity of living systems [4].
ED

There are several scientific, economical, and societal drivers like government, academia, and
industries, which promote the development and use of docking studies in various stages of
PT

the drug development process [5].With the help of docking study, thousands of molecules can
be evaluated for potential efficacy and safety at a small cost in a very short interval of time
[6]. A variety of docking approaches are currently available to predict the interaction of
E

molecules against particular receptors [7] but there is a lack of knowledge and confidence
CC

among researchers. Unfortunately, a method that is suited for certain types of prediction may
not work properly (or not work at all) for others [8]. Many compounds fail in molecular
A

dynamics (MD) simulation in-spite of possessing high doc score. Also, there is a lack of
clarity over whether the compound is an agonist or antagonist. If used correctly, docking
tools can be very effective in assessing the safety and efficacy of molecules [9]. Therefore, to
ensure accurate and effective application of docking models, it is necessary to understand the
method’s strengths, limitations, the scope of application, and interpretation; to choose the
most effective method for the problem at hand; and to customize these methods for each

3
problem, if necessary. Thus, there is need for better understanding of the docking methods. In
this review, we have tried to highlight the importance of docking methods particular in field
of Pharmacology, various computational techniques, how much accurate they are? What are
their limitations? In the future, how they could play an important role?
2. Need of Docking study in Pharmacology Field
Assessment of safety and efficacy of all categories of drugs is still mainly based on animal
experimentation, but unfortunately animal experimentation has lots of serious problems such
as question about the efficacy of drug after its long term use [10]. The chronic effects in

T
animals are impossible to study due to their small life span (4-5 years). However, in contrast,

IP
the average life of a normal human being is 70 years. Thus, results which are observed in

R
animals are difficult to extrapolate to humans [11]. To see these pharmacological effects,
animals are subjected to a much higher dose of drugs than humans would ever be exposed to

SC
in real life. This higher dose of drugs in experiments generally shows some signs of overdose,
which raises the question about accuracy and reliability. Further, Scientists extrapolate the

U
higher doses of drugs (tested on animals) to lower dose (actual exposure to human) as per
N
regulatory guidelines [12] but it’s very challenging task to get reliable results.
A
Another concern is regarding the cost of these studies. It is well known that safety and
M

efficacy study in animals is extremely expensive, or it raises moral and ethical issues because
it leads to suffer or kills many animals [13, 14]. According to the US Food and Drug
Administration (FDA), more than 90 percent promising new chemical entities (NCE) fail
ED

when they’re tested in humans, either due to efficacy or safety concerns [15]. In some of the
cases, these NCE had looked promising in various animal tests, usually in different species,
PT

but when these molecules are tested in humans, they show different effects. From the
perspective of the pharmaceutical industry, to avoid late-stage failures in the development of
E

NCE to be used as drugs, docking studies are now mostly carried out at a much earlier stage
CC

of the development process [16]. Currently, there has been a dramatic increase in the number
of active chemical entities due to the use of combinatorial chemistry libraries and high-
throughput screening (HTS)[17]. Traditional safety and efficacy methods, which test
A

chemical in animals is unable to cope with the influx of these higher numbers of new
chemical entities in a timely way.

Thus, there is growing interest in the development of alternative approaches to

pharmacological testing that reduces, refines or replaces the use of animals in safety and

4
efficacy assessment [18].Computational docking is best option to solve all the above
mentioned problems [19].

3. Molecular Docking Study

The foundation stone for powerful programs that are used to understand and predict chemical
processes was laid by Martin Karplus, Michael Levitt and Arieh Warshelin the 1970s.
And, in 2013, they were awarded the noble prize for their development of "multi-scale
methods for complex systems"[20]. More simply put, these three chemists have been

T
recognized for their development and application of methods to simulate the behaviour of

IP
molecules at various scales, from single molecules to proteins [21].

R
Docking is a method which estimates the preferred orientation of one molecule to a second

SC
when bound to each other to form a stable complex. It is one of the most frequently used
methods in structure-based drug design, due to its ability to predict the binding-conformation
of small molecule ligand to the appropriate target binding site [22].With the help of these

U
techniques, the number of molecules can be screened for biological activity at an early stage
N
of drug development. Thus, most of the pharmaceutical companies have separate department
A
for this activity where, many drugs have been designed and screened for further phase of
development [23].
M

Pharmacology is a study of pharmacokinetics & pharmacodynamics of drugs. The movement

ED

of drugs in the body is called Ph’Kinetics & Ph’Dynamics deals with action and effect
sequence as presented in figure 1 [24]. Docking study can play a major role to study kinetics
and dynamics of drugs. In kinetic studies, one can predict the interaction of drugs towards
PT

various metabolic enzymes and pharmacologists can get an idea how NCE will be
metabolized in the body.
E

After distribution, once the drug reaches the target site, it binds to particular target; it alters or
CC

modifies the existing functions of the body. The drug binds with receptor and makes a
complex as shown below. This drug-receptor complex is a central point for various
A

physiological and pharmacological effects [25]. Docking can help in predicting the affinity
and binding properties of drugs towards a particular target [26].

Generally, the best binding affinity molecule is considered to be more potent. Based on this,
the numbers of docking softwares (such as Auto Dock, Mol Dock, Maestro etc.) are available
for researchers which are user friendly and are easy to operate [27]. Thus, the numbers of

5
researchers are using these techniques and their work is published in various reputed journals.
The number of docking related papers has sharply increased over the past decades as
presented in figure 2. However, most of the papers related to docking studies have been
published by chemistry scientists and pharmacologists have published very few papers by
using these techniques. Thus, there is a need for better understanding of these techniques
among pharmacologists.
3.1. Docking algorithms
Generally, docking algorithms predict several orientations (poses) for the ligand inside the

T
biding site [28]. Several different docking algorithms have been developed as summarized in

IP
table 1 [29 to 53] and each one has their own advantages and disadvantages. These docking

R
programs make use of one or more specific search algorithms and these are the methods used
to predict the possible conformations of a binary complex. The major key point in the

SC
development of docking algorithms is the accuracy of the docking simulation [54]. The
accuracy may vary depending on what target is being tested and what kind of molecules
composes the screening library [55].
U
N
Scoring functions are mathematical approximating methods for estimating binding affinity
A
which is a main tool for lead optimization of virtual screening results, finding the highest-
M

affinity ligand against a target [56]. There are many methods available for estimating the
binding affinity between a protein and ligand from a given 3D structure of a given binary
complex. Using as input the atomic coordinates of a binary complex, through a large variety
ED

of techniques, scoring functions are able to estimate free energy of binding or binding
constant [57]. The free energy of binding is obtained with the Gibbs-Helmholtz equation:
PT

ΔG=ΔH- TΔS

ΔG is the free energy of binding, ΔH is the enthalpy, T is the temperature in Kelvin and ΔS
E

the entropy.
CC

The low binding energy of ligand protein complex indicate the high stability of complex that
means the contact time of ligand with receptor is more at that particular pose. It is
A

recommended to take the pose of ligand protein complex at this particular orientation to get
more accurate results.

3.2. Selection and Preparation of Protein

The selection of protein plays a major role in the docking results [58]. The three-dimensional
structure of particular receptor is increasing rapidly and easily available in protein data bank

6
(PDB) due to improvements in techniques for structure determination, such as high-
throughput X-ray crystallography, multidimensional NMR spectroscopy and other structural
characterization methods [59,60]. In drug discovery phase (from hit identification to lead
optimization), many of these current target proteins have been selected and screened for their
interaction with designed analogs. Most of the researchers download the protein structure
from Protein Data Bank (PDB) [61] (https://www.rcsb.org/pdb/home/home.do). All you have
to do is to add the name of receptor protein in the search option of PDB and download it.
When we add the name of receptor in search option, lots of protein structures will appear.

T
Now, the question is how will we select the structure of protein? Protein structure can be

IP
selected on the basis of X-Ray resolution and the conditions under which the protein crystal

R
structure was obtained [62].Resolution is the representation of the diffraction pattern and the
level of detail that will be seen when the electron density map is calculated. High-resolution

SC
structures (with resolution values of 1 Å or so) depict that the electrons are highly ordered
and it is easy to see every atom in the electron density map. Lower resolution structures (with

U
resolution of 3 Å or higher) represents only the basic contours of the protein chain, and the
N
atomic structure must be inferred [63]. Thus, it is recommended to select protein structure
A
with resolution less than 2 Å.
Once the protein is selected, the next question arises that; can we directly use these protein
M

structures? The answer is no, we can’t. If we use it directly, it will affect our results. After
selection, protein should be prepared by adding hydrogen ions and removing water
ED

molecules. Finally energy minimization should be done [64]. But, what if, the protein
structure is not available in PDB, then the researcher has to make the structure of protein by
PT

homology modeling. After preparation, the protein model should be validated by

Ramachandran plot or other methods, to improve the accuracy of prediction [65].
Thus, a good amount of preparatory work is needed, before performing docking studies. The
E

overall steps involved in docking have been presented in figure 3.

CC

3.3. Selection and Preparation of Ligand

The structure of ligand can be drawn in Chem draw software or can be downloaded from
A

chemistry library or zinc database. Further, energy minimization should be done before using
of these structures for docking [66]. If the protein and ligand are prepared well, then docking
can be performed.
3.4. Selection of type of docking

7
The selection of type of docking depends upon the requirement of the researcher. Docking is
mainly of two types: Rigid docking and Flexible docking [67]. In rigid docking, protein and
ligand are fixed so that bond angles or lengths are not changeable [68]. This type of docking
is extremely fast, but lacks its practical use because it neglects the conformational degrees of
freedom of ligands [69]. Flexible docking allows conformational shifts and widely used
today, although it requires much more time and computational throughput [70]. In case, if we
want to dock only a few compounds in the specific protein binding site, at a specific pH,
water, or solvation, then the flexible docking program might be a good choice but, if we need

T
to screen more compounds (in thousands) from a database, then flexible docking may not be a

IP
good choice unless and until we have a very powerful and high-speed computer [71]. A user

R
may choose different docking modes based on the ability of their computer hardware and the
character of the target protein.

SC
3.5. Selection of best docking scoring function
The selection of best docking score function depends upon the stability of ligand protein

U
complex [72]. The question arises, how can we choose the proper scoring function to obtain
N
the correct binding pose and possible ligand? This is a difficult question to answer.
A
Theoretically, a lower Gibbs free energy of a complex indicates that the protein–ligand
complex is more stable [73].
M

3.6. Validation of docking

Molecular dynamics (MD) simulations have become valuable tools in probing the details of
ED

protein structure and dynamics [74, 75]. These simulations can be used before and after
docking of protein with particular ligand to optimize its structure and account for protein
PT

flexibility; for the refinement of docked complexes, to calculate binding free energies to
include solvent effects and account for induced fit & to provide an accurate ranking of the
potential ligands. It is used to explore the stability and conformational flexibility of all the
E

protein and ligand systems. GROMACS is generally used for this purpose [76, 77].
CC

The docking mode can very roughly be divided into four models: lock and key, conformation
isomerism, induced fit and conformation selection. It seems that only the induced-fit mode
A

will change the shape of the binding pocket. However, the entry into the binding pocket will
sometimes change the shape of the binding site, according to simulation from MD
experiments [78]. When the type of shift occurs with ligand binding, rigid docking programs
will not provide useful answers. This seems to suggest that there may be errors from docking
programs that utilize a rigid protein structure.
4. Current Challenges
8
The chance of success of computational docking approaches is highly variable. It is well
known that each individual software has its own shortcomings and pitfalls. A software will
never be able to provide the same information, with the same level of reliability, as a
measurement. In silico models might not perform well if a predicted chemical is beyond the
chemical space where the softwares were developed. It is important that the softwares built
are continuously validated and refined based on new data and understanding the cause of the
apparent ‘failure’ cases. Another most significant hurdle is developing non-animal methods
that would permit assessment of the potential for a chemical or drug to cause adverse health

T
effects following repeated systemic exposure [79, 80]. Keeping all the above in mind, it is not

IP
surprising that the acceptance of these predictive tools continues to be difficult, but if the

R
existing issues can be appropriately addressed, it might eventually be well worth the effort.

SC
4.1. Accuracy of docking

As we have discussed earlier, lots of published papers related to docking are available

U
indicating their wide use in research to identify potent ligand at an early stage of the drug
N
development process. Various user friendly softwares such as Flex X, Glide, Gold, Ludi etc
are available to predict the interaction of molecules towards receptors. In literature, most of
A
the published papers have used Auto-dock and Gold to perform docking as shown in figure 4.
M

From this data, we can’t conclude that these softwares are more accurate than others. The
increased use of these software’s may be due to their free availability for the users. The use of
ED

various docking softwares in published literature from 2000 to 2017 has been complied in
figure 4. Inspite of these softwares, why it is so difficult to find a potent lead molecule for a
commercial study? Also, most of the leads showing very good binding affinity under in silico
PT

conditions but fail to show promising results when tested in animals and humans. The
accuracy of docking is questionable due to various issues [81]. First, is regarding the use of
E

protein structure. In most of the cases, protein structures are available in the protein data bank
CC

(PDB), are complexed with an internal ligand. To use these proteins, normally researchers
delete the ligand from the protein and dock their molecule. However, this process can affect
A

the accuracy of docking. The second biggest issue is the environment of the binding site. For
example, drug molecule has to bind with intracellular receptor to show its response. In this
case, even docking results show high binding under in silico environment but, when these
molecules are tested in labs, they show false results. Third issue is regarding the location of
protein which may be located at different pH, which can alter the binding of ligand with
receptors.

9
4.2. Nature of Ligand (agonist or antagonist)
The prediction of agonist or antagonistic effect of ligand through docking is not possible.
Docking study will only tell us about the binding affinity of molecule towards receptor. This
is all & nothing. However, in literature, some articles reported the agonist or antagonistic
behaviour of molecules through docking. To check this kind of behaviour, further wet lab
experiments should be performed [82, 83]. Therefore, it is advisable not to over interpret the
docking results in a paper unless other validations have been performed.

T
4.3. Docking results and MD simulation results

IP
In literature, most of the papers related to docking studies lack further validation by
molecular dynamic approaches. Docking calculate the binding affinity of ligand-protein or

R
protein-protein complex [84]. Molecular dynamics (MD) helps to refine or to increase the

SC
accuracy of docking result by studying changes of shape of complex with time, e.g. there is a
huge difference between the poses. However, it increases complexity and time per screening

U
as compared to docking experiments and the final ligand pose after MD simulations.
Sometime, in docking study, results show good binding affinity towards receptors, but after
N
MD, the ligand may not bind to pocket. Thus, to draw any conclusion from docking studies, it
A
should be validated by performing MD simulations [85]. However, the time required for
M

small simulations is one of the biggest challenges. It requires 2-3 days to complete even for
5-10ns simulation.
4.4. Regulatory Aspects
ED

Docking studies are considered to be valid & are recognized by some regulatory agencies.
Currently, there is no specific guidance regarding the use of docking softwares [86]. Thus,
PT

there is a need for regulatory guidance which will help to implement these techniques in an
effective manner.
E
CC

5. Current and Future Perspectives

In the present situation, there is commercial and consumer pressure to find and use alternative
methods for animal testing to safety and efficacy studies. Therefore an opportunity exists for
A

computational scientists to develop reliable softwares to predict behaviour of NCE.

Currently, there are various softwares and online web servers available for prediction of
behaviour of NCE [87]. However, there is lack of confidence in the accuracy of a these
methods which results in the low level of acceptance by users other than the computational
chemist and modeling groups. Thus, there is a need to be as much precise as possible and

10
modelling methods need to be chosen carefully. Further, every computer scientist must be
encouraged to work within multi-disciplined program teams, to provide the support needed
for the decision-making process in a project [88].

To create a reliable predictive model and simulations of reality is a big challenge in the field
of pharmacology due to complex physiological processes. However, greater confidence can
be obtained if one can use these techniques in a correct way. Recent advancements in the
field of bioinformation system & biology have indicated the potential of these techniques to

T
reduce the attrition rate of NCE [89]. These advancements help researchers to understand the

IP
different mechanisms of drugs which may affect the biological structure, processes and
pathways. These are some reasons that in-silico pharmacology till date & have limited

R
success in delivering in vivo relevant predictions. We hope that these facilities will speed up

SC
the development cycle of future predictive applications and will ultimately lead to improved
and more relevant applications in this area.

6. Concluding remarks
U
N
The development of NCE with potential therapeutic agents is one of the most complex and
difficult process in the pharmaceutical industry. Millions of dollars and many years are
A
devoted to the discovery of NCE. Emerging technologies indicated the role of docking
M

studies in the drug development phase. With the help of these studies, lots of NCE can be
checked for their interaction with particular receptor. However, the results of docking studies
ED

are questionable if not used in the correct way. In literature, various papers are available
which shows lack of validation of docking results by molecular dynamics simulations.
Sometimes the ligand will fly away during the MD simulation, despite a high dock score.
PT

However, practically it’s not possible to do molecular dynamic studies for every docking
study. Because of hardware and software limitations, it is currently impossible to simulate4 h
E

or 8 h of MD. Hence, we strongly suggest scientists should draw their conclusions carefully,
CC

especially if one only has the docking or virtual screening results and no further validation. In
the future, progress in either hardware or software might allow residence time to be simulated
A

by MD. However, for now, MD cannot model residence time because the simulation time is
too short (only a few milli or microseconds). Thus, there are many instances in literature
which show that a ligand with the highest dock score does not mean that the ligand is a potent
lead. That is the frequent mistake in many of the published papers. However, the field of in
silico pharmacology has been in a continuous development through the introduction of new
methods, improvement of the existing ones, or discarding of some of them. We hope that in

11
future more and more compounds will be screened at a faster pace with more accurate and
reliable results.
Acknowledgments: Authors are thankful to Mr. Parveen Garg, Chairman, ISFCP, for
providing research facility.

Declaration of Interest statement

The authors declare no conflict of intersect.

T
References

IP
1. Kitchen, D.B., Decornez, H., Furr, J.R. and Bajorath, J., 2004. Docking and scoring in
virtual screening for drug discovery: methods and applications. Nature reviews. Drug

R
discovery, 3(11), p.935.
2. Kavlock, R.J., Ankley, G., Blancato, J., Breen, M., Conolly, R., Dix, D., Houck, K.,

SC
Hubal, E., Judson, R., Rabinowitz, J. and Richard, A., 2007. Computational
toxicology—a state of the science mini review. Toxicological Sciences, 103(1), pp.14-
27.

U
3. Sliwoski, G., Kothiwale, S., Meiler, J. and Lowe, E.W., 2014. Computational methods
N
in drug discovery. Pharmacological reviews, 66(1), pp.334-395.
4. Ekins, S., Mestres, J. and Testa, B., 2007. In silico pharmacology for drug discovery:
A
methods for virtual ligand screening and profiling. British journal of
pharmacology, 152(1), pp.9-20.
M

5. Chen, Y.Z. and Zhi, D.G., 2001. Ligand–protein inverse docking and its potential use
in the computer search of protein targets of a small molecule. Proteins: Structure,
Function, and Bioinformatics, 43(2), pp.217-226.
ED

6. Festing, M.F., 1994. Reduction of animal use: experimental design and quality of
experiments. Laboratory Animals, 28(3), pp.212-221.
7. Hou, T. and Xu, X., 2004. Recent development and application of virtual screening in
PT

drug discovery: an overview. Current pharmaceutical design, 10(9), pp.1011-1033.

8. Abagyan, R. and Totrov, M., 2001. High-throughput docking for lead
generation. Current opinion in chemical biology, 5(4), pp.375-382.
E

9. Klebe, G., 2006. Virtual ligand screening: strategies, perspectives and

CC

limitations. Drug discovery today, 11(13), pp.580-594.

10. Doke, S.K. and Dhawale, S.C., 2015. Alternatives to animal testing: A review. Saudi
Pharmaceutical Journal, 23(3), pp.223-229.
A

11. Esmon, C.T., 2004. Why do animal models (sometimes) fail to mimic human
sepsis?. Critical care medicine, 32(5), pp.S219-S222.
12. Dayan, A.D., 1981. The relative worth of animal testing. In Risk-Benefit Analysis in
Drug Research (pp. 97-112). Springer Netherlands.
13. Lipinski, C.A., Lombardo, F., Dominy, B.W. and Feeney, P.J., 2001. Experimental
and computational approaches to estimate solubility and permeability in drug

12
 discovery and development settings1. Advanced drug delivery reviews, 46(1-3), pp.3-
26.
14. Latin, H., 1988. Good science, bad regulation, and toxic risk assessment. Yale J. on
Reg., 5, p.89.
15. Burke, S.P., Stratton, K. and Baciu, A. eds., 2007. The future of drug safety:
promoting and protecting the health of the public. National Academies Press.
16. Smith, G.R. and Sternberg, M.J., 2002. Prediction of protein–protein interactions by
docking methods. Current opinion in structural biology, 12(1), pp.28-35.
17. Bleicher, K.H., Böhm, H.J., Müller, K. and Alanine, A.I., 2003. Hit and lead
generation: beyond high-throughput screening. Nature reviews. Drug discovery, 2(5),

T
p.369.

IP
18. Balls, M., 1994. Replacement of animal procedures: alternatives in research,
education and testing. Laboratory Animals, 28(3), pp.193-211.

R
19. Fogel, D.B., 2000. What is evolutionary computation?. IEEE Spectrum, 37(2), pp.26-
32.

SC
20. Fersht, A.R., 2013. Profile of martin karplus, michael levitt, and arieh warshel, 2013
nobel laureates in chemistry. Proceedings of the National Academy of
Sciences, 110(49), pp.19656-19657.
21. Thiel, W. and Hummer, G., 2013.
chemistry. Nature, 504(7478), p.96. U Methods for computational
N
22. Gohlke, H., Hendlich, M. and Klebe, G., 2000. Knowledge-based scoring function to
A
predict protein-ligand interactions. Journal of molecular biology, 295(2), pp.337-356.
23. Leach, A.R., Shoichet, B.K. and Peishoff, C.E., 2006. Prediction of protein− ligand
M

interactions. Docking and scoring: Successes and gaps. Journal of medicinal

chemistry, 49(20), pp.5851-5855.
24. Ghosh, M., 2007. Fundamentals of experimental pharmacology. Indian Journal of
ED

Pharmacology, 39(4), pp.216-216.

25. Julien, R.M., 2013. A primer of drug action: A concise nontechnical guide to the
actions, uses, and side effects of psychoactive drugs, revised and updated. Holt
PT

Paperbacks.
26. Klebe, G., 2000. Recent developments in structure-based drug design. Journal of
Molecular Medicine, 78(5), pp.269-281.
E

27. Baldi, A., 2010. Computational approaches for drug design and discovery: An
CC

overview. Systematic Reviews in Pharmacy, 1(1), p.99.

28. Kontoyianni, M., McClellan, L.M. and Sokol, G.S., 2004. Evaluation of docking
performance: comparative data on docking algorithms. Journal of medicinal
A

chemistry, 47(3), pp.558-565.

29. Baldi, A., 2010. Computational approaches for drug design and discovery: An
overview. Systematic Reviews in Pharmacy, 1(1), p.99.
30. Trott, O. and Olson, A.J., 2010. AutoDock Vina: improving the speed and accuracy of
docking with a new scoring function, efficient optimization, and
multithreading. Journal of computational chemistry, 31(2), pp.455-461.

13
31. Dolle, R.E., 1997. Comprehensive survey of chemical libraries yielding enzyme
inhibitors, receptor agonists and antagonists, and other biologically active agents:
1992 through 1997. Molecular diversity, 3(4), pp.199-233.
32. Zsoldos, Z., Reid, D., Simon, A., Sadjad, S.B. and Johnson, A.P., 2007. eHiTS: a new
fast, exhaustive flexible ligand docking system. Journal of Molecular Graphics and
Modelling, 26(1), pp.198-212.
33. Baldi, A., 2010. Computational approaches for drug design and discovery: An
overview. Systematic Reviews in Pharmacy, 1(1), p.99.
34. Akula, Nagaraju, et al. "3D QSAR studies of AChE inhibitors based on molecular
docking scores and CoMFA." Bioorganic & medicinal chemistry letters 16.24 (2006):

T
6277-6280.

IP
35. Kramer, Bernd, Matthias Rarey, and Thomas Lengauer. "Evaluation of the FLEXX
incremental construction algorithm for protein–ligand docking." Proteins: Structure,

R
Function, and Bioinformatics 37.2 (1999): 228-241.
36. Halgren, T.A., Murphy, R.B., Friesner, R.A., Beard, H.S., Frye, L.L., Pollard, W.T.

SC
and Banks, J.L., 2004. Glide: a new approach for rapid, accurate docking and scoring.
2. Enrichment factors in database screening. Journal of medicinal chemistry, 47(7),
pp.1750-1759.

U
37. Verdonk, M.L., Cole, J.C., Hartshorn, M.J., Murray, C.W. and Taylor, R.D., 2003.
Improved protein–ligand docking using GOLD. Proteins: Structure, Function, and
N
Bioinformatics, 52(4), pp.609-623.
A
38. Meng, E.C., Kuntz, I.D., Abraham, D.J. and Kellogg, G.E., 1994. Evaluating docked
complexes with the HINT exponential function and empirical atomic
M

hydrophobicities. Journal of computer-aided molecular design, 8(3), pp.299-306.

39. Biesiada, J., Porollo, A., Velayutham, P., Kouril, M. and Meller, J., 2011. Survey of
public domain software for docking simulations and virtual screening. Human
ED

genomics, 5(5), p.497.

40. Wriggers, W., Milligan, R.A. and McCammon, J.A., 1999. Situs: a package for
docking crystal structures into low-resolution maps from electron
PT

microscopy. Journal of structural biology, 125(2-3), pp.185-195.

41. Baldi, A., 2010. Computational approaches for drug design and discovery: An
overview. Systematic Reviews in Pharmacy, 1(1), p.99.
E

42. Michino, M., Abola, E., Dock, G.P.C.R., Brooks, C.L., Dixon, J.S., Moult, J. and
CC

Stevens, R.C., 2009. Community-wide assessment of GPCR structure modelling and

ligand docking: GPCR Dock 2008. Nature Reviews Drug Discovery, 8(6), pp.455-
463.
A

43. Abagyan, R., Totrov, M. and Kuznetsov, D., 1994. ICM—a new method for protein
modeling and design: applications to docking and structure prediction from the
distorted native conformation. Journal of computational chemistry, 15(5), pp.488-
506.
44. Tovchigrechko, A. and Vakser, I.A., 2006. GRAMM-X public web server for
protein–protein docking. Nucleic acids research, 34(suppl_2), pp.W310-W314.

14
45. Ackermann, F., Herrmann, G., Posch, S. and Sagerer, G., 1998. Estimation and
filtering of potential protein-protein docking positions. Bioinformatics (Oxford,
England), 14(2), pp.196-205.
46. Palma, P.N., Krippahl, L., Wampler, J.E. and Moura, J.J., 2000. BiGGER: a new
(soft) docking algorithm for predicting protein interactions. Proteins: Structure,
Function, and Bioinformatics, 39(4), pp.372-384.
47. Comeau, S.R., Gatchell, D.W., Vajda, S. and Camacho, C.J., 2004. ClusPro: a fully
automated algorithm for protein–protein docking. Nucleic acids
research, 32(suppl_2), pp.W96-W99.
48. Böhm, H.J., 1994. On the use of LUDI to search the Fine Chemicals Directory for

T
ligands of proteins of known three-dimensional structure. Journal of computer-aided

IP
molecular design, 8(5), pp.623-632.
49. Magesh, S., Savita, V., Moriya, S., Suzuki, T., Miyagi, T., Ishida, H. and Kiso, M.,

R
2009. Human sialidase inhibitors: design, synthesis, and biological evaluation of 4-
acetamido-5-acylamido-2-fluoro benzoic acids. Bioorganic & medicinal

SC
chemistry, 17(13), pp.4595-4603.
50. Ponomarenko, J., Bui, H.H., Li, W., Fusseder, N., Bourne, P.E., Sette, A. and Peters,
B., 2008. ElliPro: a new structure-based tool for the prediction of antibody
epitopes. BMC bioinformatics, 9(1), p.514.
U
51. van Dijk, M., van Dijk, A.D., Hsu, V., Boelens, R. and Bonvin, A.M., 2006.
N
Information-driven protein–DNA docking using HADDOCK: it is a matter of
A
flexibility. Nucleic acids research, 34(11), pp.3317-3325.
52. Ritchie, D.W., 2003. Evaluation of protein docking predictions using Hex 3.1 in
M

CAPRI rounds 1 and 2. Proteins: Structure, Function, and Bioinformatics, 52(1),

pp.98-106.
53. Baldi, A., 2010. Computational approaches for drug design and discovery: An
ED

overview. Systematic Reviews in Pharmacy, 1(1), p.99

54. Dias, R., de Azevedo, J. and Walter, F., 2008. Molecular docking algorithms. Current
Drug Targets, 9(12), pp.1040-1047.
PT

55. Kellenberger, E., Rodrigo, J., Muller, P. and Rognan, D., 2004. Comparative
evaluation of eight docking tools for docking and virtual screening
accuracy. Proteins: Structure, Function, and Bioinformatics, 57(2), pp.225-242.
E

56. Warren, G.L., Andrews, C.W., Capelli, A.M., Clarke, B., LaLonde, J., Lambert,
CC

M.H., Lindvall, M., Nevins, N., Semus, S.F., Senger, S. and Tedesco, G., 2006. A
critical assessment of docking programs and scoring functions. Journal of medicinal
chemistry, 49(20), pp.5912-5931.
A

57. Beveridge, D.L. and DiCapua, F.M., 1989. Free energy via molecular simulation:
applications to chemical and biomolecular systems. Annual review of biophysics and
biophysical chemistry, 18(1), pp.431-492.
58. Chandrika, B.R., Subramanian, J. and Sharma, S.D., 2009. Managing protein
flexibility in docking and its applications. Drug discovery today, 14(7), pp.394-400.
59. Blundell, T.L., Jhoti, H. and Abell, C., 2002. High-throughput crystallography for
lead discovery in drug design. Nature reviews. Drug discovery, 1(1), p.45.

15
60. Vranken, W.F., Boucher, W., Stevens, T.J., Fogh, R.H., Pajon, A., Llinas, M., Ulrich,
E.L., Markley, J.L., Ionides, J. and Laue, E.D., 2005. The CCPN data model for NMR
spectroscopy: development of a software pipeline. Proteins: Structure, Function, and
Bioinformatics, 59(4), pp.687-696.
61. Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T.N., Weissig, H.,
Shindyalov, I.N. and Bourne, P.E., 2006. The Protein Data Bank, 1999–.
In International Tables for Crystallography Volume F: Crystallography of biological
macromolecules (pp. 675-684). Springer Netherlands.
62. Creighton, T.E., 1993. Proteins: structures and molecular properties. Macmillan.
63. Berman, H.M., Battistuz, T., Bhat, T.N., Bluhm, W.F., Bourne, P.E., Burkhardt, K.,

T
Feng, Z., Gilliland, G.L., Iype, L., Jain, S. and Fagan, P., 2002. The protein data

IP
bank. Acta Crystallographica Section D: Biological Crystallography, 58(6), pp.899-
907.

R
64. Kramer, B., Rarey, M. and Lengauer, T., 1999. Evaluation of the FLEXX incremental
construction algorithm for protein–ligand docking. Proteins: Structure, Function, and

SC
Bioinformatics, 37(2), pp.228-241.
65. Hooft, R.W., Sander, C. and Vriend, G., 1997. Objectively judging the quality of a
protein structure from a Ramachandran plot. Bioinformatics, 13(4), pp.425-430.

U
66. Andricopulo, A.D., Guido, R.V. and Oliva, G., 2008. Virtual screening and its
integration with modern drug design technologies. Current medicinal
N
chemistry, 15(1), pp.37-46.
A
67. Gehlhaar, D.K., Verkhivker, G.M., Rejto, P.A., Sherman, C.J., Fogel, D.R., Fogel,
L.J. and Freer, S.T., 1995. Molecular recognition of the inhibitor AG-1343 by HIV-1
M

protease: conformationally flexible docking by evolutionary programming. Chemistry

& biology, 2(5), pp.317-324.
68. Chen, Y.C., 2015. Beware of docking!. Trends in pharmacological sciences, 36(2),
ED

pp.78-95.
69. Sousa, S.F., Fernandes, P.A. and Ramos, M.J., 2006. Protein–ligand docking: current
status and future challenges. Proteins: Structure, Function, and Bioinformatics, 65(1),
PT

pp.15-26.
70. Kitchen, D.B., Decornez, H., Furr, J.R. and Bajorath, J., 2004. Docking and scoring in
virtual screening for drug discovery: methods and applications. Nature reviews. Drug
E

discovery, 3(11), p.935.

CC

71. Graves, A.P., Brenk, R. and Shoichet, B.K., 2005. Decoys for docking. Journal of
medicinal chemistry, 48(11), pp.3714-3728.
72. Gohlke, H., Hendlich, M. and Klebe, G., 2000. Knowledge-based scoring function to
A

predict protein-ligand interactions. Journal of molecular biology, 295(2), pp.337-356.

73. Krissinel, E. and Henrick, K., 2007. Inference of macromolecular assemblies from
crystalline state. Journal of molecular biology, 372(3), pp.774-797.
74. Klepeis, J.L., Lindorff-Larsen, K., Dror, R.O. and Shaw, D.E., 2009. Long-timescale
molecular dynamics simulations of protein structure and function. Current opinion in
structural biology, 19(2), pp.120-127.
75. Schlick, T., 2010. Molecular modeling and simulation: an interdisciplinary guide: an
interdisciplinary guide (Vol. 21). Springer Science & Business Media.

16
76. Van Der Spoel, D., Lindahl, E., Hess, B., Groenhof, G., Mark, A.E. and Berendsen,
H.J., 2005. GROMACS: fast, flexible, and free. Journal of computational
chemistry, 26(16), pp.1701-1718.
77. Berendsen, H.J., van der Spoel, D. and van Drunen, R., 1995. GROMACS: a
message-passing parallel molecular dynamics implementation. Computer Physics
Communications, 91(1-3), pp.43-56.
78. Rajamani, D., Thiel, S., Vajda, S. and Camacho, C.J., 2004. Anchor residues in
protein–protein interactions. Proceedings of the National Academy of Sciences of the
United States of America, 101(31), pp.11287-11292.
79. Sousa, S.F., Fernandes, P.A. and Ramos, M.J., 2006. Protein–ligand docking: current

T
status and future challenges. Proteins: Structure, Function, and Bioinformatics, 65(1),

IP
pp.15-26.
80. Basketter, D.A., Clewell, H., Kimber, I., Rossi, A., Blaauboer, B.J., Burrier, R.,

R
Daneshian, M., Eskes, C., Goldberg, A. and Hasiwa, N., 2012. A roadmap for the
development of alternative (non-animal) methods for systemic toxicity testing-t4

SC
report. ALTEX-Alternatives to Animal Experimentation, 29(1), pp.3-91.
81. Hou, T. and Xu, X., 2004. Recent development and application of virtual screening in
drug discovery: an overview. Current pharmaceutical design, 10(9), pp.1011-1033.

U
82. Moro, S., Guo, D., Camaioni, E., Boyer, J.L., Harden, T.K. and Jacobson, K.A., 1998.
Human P2Y1 receptor: molecular modeling and site-directed mutagenesis as tools to
N
identify agonist and antagonist recognition sites. Journal of medicinal
A
chemistry, 41(9), pp.1456-1466.
83. Prinz, F., Schlange, T. and Asadullah, K., 2011. Believe it or not: how much can we
M

rely on published data on potential drug targets?. Nature reviews Drug

discovery, 10(9), pp.712-712.
84. Gilson, M.K. and Zhou, H.X., 2007. Calculation of protein-ligand binding
ED

affinities. Annual review of biophysics and biomolecular structure, 36.

85. Rapaport, D.C., Blumberg, R.L., McKay, S.R. and Christian, W., 1996. The art of
molecular dynamics simulation. Computers in Physics, 10(5), pp.456-456.
PT

86. Walters, W.P., Stahl, M.T. and Murcko, M.A., 1998. Virtual screening—an
overview. Drug Discovery Today, 3(4), pp.160-178.
87. Singla, D., Dhanda, S.K., Chauhan, J.S., Bhardwaj, A., Brahmachari, S.K. and
E

Raghava, G.P., 2013. Open source software and web services for designing
CC

therapeutic molecules. Current topics in medicinal chemistry, 13(10), pp.1172-1191.

88. Modi, S., Hughes, M., Garrow, A. and White, A., 2012. The value of in silico
chemistry in the safety assessment of chemicals in the consumer goods and
A

pharmaceutical industries. Drug discovery today, 17(3), pp.135-142.

89. Comeau, S.R., Gatchell, D.W., Vajda, S. and Camacho, C.J., 2004. ClusPro: an
automated docking and discrimination method for the prediction of protein
complexes. Bioinformatics, 20(1), pp.45-50.

17
Figure Captions
Figure 1: Overall role of docking study in field of Pharmacology. Structure has been drawn in MS
Power point.
Figure 2: Number of papers published from year 2000 to 2017. Data has been extracted from Pub
med. Figure has been created by using MS excel.
Figure 3: Overall docking procedure. Figure has been drawn in MS Power point.
Figure 4: Use of docking softwares in published papers from 2000-2017. Pie chart has been drawn in

T
MS Excel.

R IP
SC
U
N
A
M
ED
E PT
CC

Figure 1
A

18
 A
CC
EPT
ED

19
M

Figure 2
A
N
U
SC
RIP
T
 T
RIP
SC
U
N
A
Figure 3
M
ED
EPT
CC
A

Figure 4

20
Table 1: Softwares and Webserver for docking

Sr. No. Name of Features References

software
s
1. Affinity The energy of the ligand/receptor complex is [Baldi, 2010]
utilised to automaticallyfind the best binding
modes.
2. AutoDoc Docking of the ligand is performed to a set of grids [Goodsell et al., 1996]
k describing the target protein.

T
AutoGrid Pre-calculates the grids.
AutoTors Find out which bonds will be treated as rotatablein

IP
the ligand.
3. Combibui Screen a library to predict which ones will be the [Dolle, 1997]
ld most potent.

R
4. DockVisi Includes Monte Carlo, Genetic Algorithm and [Thompson et al., 1999]
on databasescreening docking algorithms.

SC
5. FRED Examines all possible poses within a protein active [Baldi, 2010]
site, sorting for shape complementarity and

6. FlexiDoc U
optional pharmacaphoricfeatures before scoring
with more conventional functions.
Flexible docking of ligands into binding sites on [Akula et al., 2006]
N
k proteins & fast genetic algorithm for generation of
configurations.
A
7. FlexX Predicts protein-ligand Interactions. [Kramer et al., 1999]
Two main applications: Complex prediction &
M

Virtual screening.
8. Glide Identify the best binding mode through Monte [Halgren et al., 2004]
Carlo sampling.
ED

An accurate scoring function for ranking of

bindingaffinities.
9. Gold Calculates docking modes of small molecules into [Verdonk et al., 2003]
protein binding sites &Studies full ligand and
PT

partial protein flexibility.

10. Hint Empirical molecular modeling system with new [Meng et al., 1994]
methods for de novo drug design and protein or
nucleic acid structuralanalysis.
E

Evaluates/predicts effects of site-directed

mutagenesis on protein structure and stability.
CC

11. Ligplot Automatically plots protein-ligand interactions. [Wallace et al., 1995]

Generates schematic diagrams of protein-ligand
interactions for a given PDB file.
12. Situs Program package for modeling of atomic [Wriggers et al., 1999]
A

resolution structures into low-resolution density

maps; supports both rigid-body and flexible
docking
13. Vega Calculates ligand-receptor interaction energy. [Baldi, 2010]
14. Dock Generates many possible orientations of a ligand
within a user-selected regionof a receptor [Michino et al., 2009]
structure.
Evaluates likely orientations of a single ligand, or
to rank molecules from a database.

21
15. Icm-Dock Provides access to the chemical information and a [Abagyan et al., 1994]
unique set of tools for accurate ligand-protein
docking, peptide-protein docking and protein-
protein docking.
16. GRAMM Empirical approach to smoothen the [Toychigrechko and
intermolecular energy functions by changing the Vasker, 2006]
range of the atom-atom potentials.
Performs an exhaustive six-dimensional search
through the relative translations and rotations of
the molecules.
Used for protein-protein and protein-ligand
docking.

T
17. Bielefeld Detects geometrical and chemical [Ackermann et al., 1998]
Protein complementarities between surfaces of proteins

IP
Docking and estimates docking positions.
18. Bigger Predicts the structure of binary protein complexes [Palma et al., 2000]
from theunbound structures.

R
Search the complete binding space and select a set
of candidatecomplexes.

SC
Evaluates and rank each candidate according to the
estimatedprobability of being an accurate model of
the native complex.
19. ClusPro
U
Integrated approach for protein-protein docking. [Comeau et al., 2004]
N
20. Ludi Fits molecules into the active site of a receptor by [Bohm, 1994]
matching complementary polar and hydrophobic
groups &Suggests modifications to increase the
A
binding affinity of ligand.
21. Ludi/CA Calculates molecular interaction sites on a receptor [Baldi, 2010]
M

P to search suitable ligands.

Eliminates redundant hits.
22. Dot Used for computation of the electrostatic potential [Baldi, 2010]
ED

[Daughter energy between two proteins or other charged

Of molecules.
TURNIP]
23. Haddock High-Ambiguity Driven protein-protein Docking. [Van Djik et al., 2006]
PT

Generates biochemical and/or biophysical

interaction datasuch as chemical shift perturbation
data resulting from nuclearmagnetic resonance
titration experiments or mutagenesisdata
E

introduced as ambiguous interaction restraints to

drive thedocking process.
CC

24. Hex Protein docking and molecular superposition [Ritchie, 2003]

program& uses spherical polar Fourier correlations
to accelerate dockingCalculations.
25. Rachel Designed to optimize weak binding lead [Baldi, 2010]
A

compounds in an automated, combinatorial

fashion.
Allows the user to easily generate focused scoring
functions to estimate ligand binding to a specific
target receptor.

22