Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery

Abstract

Molecular docking is a structure-based computational method that generates the binding pose and affinity between ligands and targets. There are many powerful docking programs. However, there is no single program that is suitable for every system. Hence, an appropriate program is chosen based on availability, need, and computer capacity. Molecular docking has clear steps that should be followed carefully to get a good result.

Molecular docking has many applications at various stages in drug discovery. Although it has various application areas, it is commonly applied in virtual screening and drug repurposing. As a result, it is playing a substantial role in the endeavor to discover a potent drug against COVID-19. There are also approved drugs in the pharmaceutical market that are developed through the use of molecular docking. As the accessible data is increasing and the method is advancing with the contribution of the latest computational developments, its use in drug discovery is also increasing.

Molecular docking has played a crucial role in making drug discovery faster, cheaper, and more effective. More advances in docking algorithms, integration with other computational methods, and the introduction of new approaches are expected. Thus, more applications that will make drug discovery easier are expected.

Full text

Send Orders for Reprints to reprints@benthamscience.net
Letters in Drug Design & Discovery, XXXX, XX, 1-16 1
REVIEW ARTICLE
1570-1808/XX $65.00+.00 © XXXX Bentham Science Publishers
Molecular Docking: Principles, Advances, and Its Applications in Drug
Discovery
Muhammed Tilahun Muhammed1,* and Esin Aki-Yalcin2
1Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Suleyman Demirel University, Isparta, Turkey;
2Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Cyprus Health and Social Sciences University,
Guzelyurt, Northern Cyprus
Abstract: Molecular docking is a structure-based computational method that generates the binding pose
and affinity between ligands and targets. There are many powerful docking programs. However, there is
no single program that is suitable for every system. Hence, an appropriate program is chosen based on
availability, need, and computer capacity. Molecular docking has clear steps that should be followed care-
fully to get a good result.
Molecular docking has many applications at various stages in drug discovery. Although it has various
application areas, it is commonly applied in virtual screening and drug repurposing. As a result, it is play-
ing a substantial role in the endeavor to discover a potent drug against COVID-19. There are also ap-
proved drugs in the pharmaceutical market that are developed through the use of molecular docking. As
the accessible data is increasing and the method is advancing with the contribution of the latest computa-
tional developments, its use in drug discovery is also increasing.
Molecular docking has played a crucial role in making drug discovery faster, cheaper, and more effective.
More advances in docking algorithms, integration with other computational methods, and the introduction of
new approaches are expected. Thus, more applications that will make drug discovery easier are expected.
A R T I C L E H I S T O R Y
Received: April 28, 2022
Revised: July 22, 2022
Accepted: August 18, 2022
DOI:
10.2174/1570180819666220922103109
Keywords: CADD, computational method, drug design, drug discovery, molecular docking, molecular modeling.
1. INTRODUCTION
Computer-aided drug design (CADD) is an area that con-
sists of many computational strategies for the discovery,
design, and development of novel therapeutic agents. CADD
has a crucial role in improving active ligands, discovering
novel drugs and understanding biological processes at a mo-
lecular level [1]. Furthermore, the application areas of
CADD methods are widening with the increase in biological
and chemical data, increase in data storage capacity, increase
in identified drug targets, and advance in data processing
capacity [2].
CADD methods enable rapid, economic, and more effi-
cient drug discovery and development [3]. The drug devel-
opment process includes drug discovery, preclinical studies,
clinical phase studies, and registration. This is an expensive
process that takes more than 10 years on average [4]. CADD
has applications mainly in the drug discovery phase of this
process [5]. CADD provides the advantage of filtering
smaller series of compounds expected to be active from large
compound libraries and therefore guiding to find of the lead
*Address correspondence to this author at the Department of Pharmaceuti-
cal Chemistry, Faculty of Pharmacy, Suleyman Demirel University, Isparta,
Turkey; E-mail: muh.tila@gmail.com
compounds, optimization of the lead compound, and design-
ing novel compounds in the drug discovery [6, 7]. In addi-
tion to this, CADD methods can sometimes replace in vivo
models and lead to the formation of high-quality datasets [8].
There are several approved drugs in the market that are de-
veloped through CADD. For example, the anti-HIVs ralte-
gravir, saquinavir, indinavir, and ritonavir, the anti-influenza
oseltamivir, the antihypertensive captopril, the carbonic an-
hydrase inhibitor dorzolamide and the neuraminidase inhibi-
tor zanamivir are developed by using CADD methods [9].
Based on the type of data available, computer-aided drug
design methods can be categorized as target-based and lig-
and-based drug design methods [10]. In the target-based
drug design, the aim is to design potential active compounds
by using target structures. Although docking is a typical ex-
ample of this approach, molecular dynamics and binding site
estimation methods can also be evaluated in this class [11,
12]. Homology modeling can be employed at this approach's
early stages in case the protein's three-dimensional structure
hasn’t been determined yet [13]. In the ligand-based drug
design, the aim is to interpret the structure of the target by
using the structure of active ligands. Pharmacophore model-
ing and QSAR (quantitative structure-activity relationships)
are examples of this approach [14]. QSAR models effective-
ly estimate experimental activities based on molecular de-

2 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
scriptors [15]. To increase the performance of CADD, it is
important to use both approaches together in a way that
complements each other. This is also known as the hybrid
approach [1]. For example, molecular dynamics is utilized in
both methods for the discovery of novel drug candidates
[13]. There is also a method named de novo molecular de-
sign that is used to design novel chemical entities which sat-
isfy a desired molecular profile [16]. This method gives the
opportunity to generate novel molecular structures in the
abscence of a starting template [17]. In this study, molecular
docking, which is one of the target-based drug design meth-
ods, is reviewed.
Molecular docking is a structure-based computational
method that generates the binding mode and affinity between
ligands and targets by predicting their interactions [18, 19].
There are several docking tools used for this purpose. Auto-
Dock, AutoDock Vina, GOLD, Glide, MOE, ICM, and
FlexX are amongst the popular software in use [20]. Many of
them are powerful docking tools, but there is no sole soft-
ware suitable for every system. Thus, users should choose
their preferable software based on availability, their needs,
and their computer capacity. It is also possible to use more
than one software in a way that increases the quality of the
output [21].
Molecular docking has various applications in the drug
discovery and design process. In the early years of its appli-
cation, it was mainly used to investigate the molecular inter-
actions between ligands and targets [22]. Nowadays, it sup-
ports wider and more diverse areas of drug discovery [23]. It
has applications in virtual screening, target fishing, drug side
effect prediction, polypharmacology, and drug repurposing.
With the involvement of state-of-the-art computational ap-
proaches like artificial intelligence, there is an advance in
docking algorithms. The integration of molecular docking
with other approaches, such as ligand-based methods, is also
underway. Moreover, there is a significant increase in open-
source biological and chemical data. All these are contrib-
uting to the advance in molecular docking. With the advanc-
es in molecular docking, its application in drug discovery is
rapidly increasing [11, 24].
Molecular docking has substantially brought anti-HIV,
anticancer, and various other drugs to the pharmaceutical
market [25]. A typical example of the successful application
of molecular docking is the design of rilpivirine [26]. The
molecular modeling studies that led to the discovery of rilpi-
virine involved the docking of diarylpyrimidine ligands into
the reverse transcriptase binding site. The computational
assessment followed by the experimental evaluation resulted
in the approval of rilpivirine against HIV [27]. Similarly,
molecular docking guided the design of betrixaban. A lead
compound with improved potency was discovered [28]. The
binding mode of the lead compound was elucidated by dock-
ing using the GOLD program. Based on docking and other
computational observations, further modifications that led to
the development of betrixaban were performed [29]. In an-
other example, molecular docking was used in the discovery
of the neuraminidase inhibitor zanamivir. Molecular docking
was used to analyze the active site of neuraminidase and its
interactions with its new inhibitors. The interaction of
zanamivir with neuraminidase was elucidated [30]. After the
computational results were confirmed by the in vivo tests, the
drug was approved [25]. Flexible molecular docking was
also used in the discovery and development of drugs [31].
For instance, it was utilized in the discovery of vaborbactam.
ICM docking was used in the design of lead β-lactamase
enzyme inhibitors with better activity. The lead molecule
was evaluated with computational and experimental meth-
ods. Finally, vaborbactam was developed [32].
There is a great effort worldwide to discover an effective
drug to combat the global pandemic, COVID-19 (Corona-
virus disease 2019) [33]. The available literature shows the
discovery of new potent molecules against novel coronavirus
is still at its early stage [34]. Thus, virtual screening and drug
repurposing are recommended as the fastest options for the
discovery of a potent drug against the novel coronavirus [35,
36]. As molecular docking has been commonly used in vir-
tual screening and drug repurposing, it can play a substantial
role in discovering a potent drug against the novel corona-
virus [37]. Therefore, it has been applied and recommended
to be used in the endeavor to discover promising drugs
against the coronavirus using computational methods [38].
Molecular docking has established applications at various
stages of the drug discovery process. With the contribution
of the latest computational advances, it is expected to have
more applications [25]. Thus, updated information about
molecular docking is in need. This review is aimed at meet-
ing this demand in academia and the pharmaceutical indus-
try. In this work, the basic principles of molecular docking,
including its steps are presented. Current applications of mo-
lecular docking in drug discovery are explained with exam-
ples. Furthermore, the challenges and advances in molecular
docking are summarized.
2. PRINCIPLES OF MOLECULAR DOCKING
Docking is a method based on the examination of the
fitting of the designed compounds to target cavities and their
interactions with the residues [39]. In the computational drug
discovery process, docking is generally undertaken between
small molecules and macromolecules, as in protein-ligand
docking. This type of docking is known as molecular dock-
ing. Over the last few years, docking has also been per-
formed between two macromolecules, as in protein-protein
docking [40].
The basis for the majority of the docking programs is
molecular mechanics, which explains polyatomic systems
using classical physics. Experimental parameters are used to
reduce the deviation between the experimental data and mo-
lecular mechanics. Due to the limitations of the experimental
methods, mathematical equations are converted into parame-
ters using quantum mechanics semiempirical and ab initio
theoretical calculations [39]. In this regard, it is a set of equa-
tions with different parameters that aim to define molecular
force field systems, which are based on potential energy,
torsional properties, the geometry of the bond, electrostatic
terms, and Lenard-Jones potential. AMBER, CHARMM,
GROMOS, OPLS-AA, and UFF have known examples of
force fields [41].

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 3
In the 1980s, molecular modeling was performed using
force fields. In the continuation of these methods, modeling
of molecular processes like the binding of ligands to their
target proteins was undertaken. Two main methods are built
for this purpose: Rigid body and flexible docking [10]. In
rigid body docking, ligands and targets are considered two
different bodies that recognize each other according to their
shape and size. In flexible docking, protein-ligand recogni-
tion occurs by considering the effect of the two structures on
each other [42]. In early practices, a rigid ligand was docked
into a rigid target. With the advances in computing power,
new efficient computational methods that enable the docking
of flexible ligands into rigid targets are introduced. As the
targets are also flexible at physiological conditions, their
conformational changes are expected to be addressed. Oth-
erwise, ligands that could bind to a target could give a mis-
taken interaction in computational analysis. This is addressed
by introducing target flexibility in molecular docking. Ad-
dressing the targeting flexibility requires additional computa-
tional resources. Approximate methods that make it practical
have been introduced [43].
There are many servers and programs that are used in
molecular docking. In each program, various force fields and
algorithms are used for pose prediction, refinement, and gen-
eration of the target-ligand interactions (Table 1) [44]. Alt-
hough there are many powerful docking programs, it is good
to remember that none of the docking algorithms in use are
suitable for every system. It is recommended to use more
than one program [21].
2.1. General Recommendations and Guidelines for Mo-
lecular Docking
2.1.1. Hardware and Software Requirements for Molecular
Docking
As ligand docking and computing are performed in a few
minutes, docking computations are not considered intensive
processing unit (CPU). Currently, almost every personal
computer (PC) is capable of running small docking works
(500-1000 compounds) in an acceptable time [61]. However,
in the virtual screening of public databases using docking-
based methods, the number of molecules could rise rapidly
(106 compounds). This requires more data processors to fin-
ish the process in a reasonable time. Generally, GPU data
processing is more efficient and attractive for intensive pro-
cessing than CPU-based computations [40].
2.1.2. Program Selection in Molecular Docking
There are many docking methods and approaches (Table
1). Among these methods, for beginners, easily accessible
academic or free software are preferable. Some of the dock-
ing programs are not designed to run on Windows [62]. In
such cases, beginners can start with Linux and overcome the
problem. In addition to this, Windows-friendly programs
such as AutoDock, Vina, and LeDock can be preferred [40].
2.2. Steps of Molecular Docking
The molecular docking process consists of target protein
determination and preparation, ligand preparation, determi-
nation of the type of docking to be used, selection of the best
docking scoring function, and validation (Fig. 1) [40, 63].
2.2.1. Target Protein Determination and Preparation
The properties of the selected protein structure affect
docking results [64]. With the development of X-ray crystal-
lography, NMR, Cryo-EM, and similar structure determina-
tion methods, the number of proteins with known three-
dimensional (3D) structures is rapidly increasing and they
are accessible to the public in databases like the protein data
bank (PDB) [65]. The first step of docking is retrieving the
3D structure of the protein, preferably bound by a ligand,
from the PDB. Using 3D structures with high resolution (˂
2Å) or structures bound by a high-affinity ligand is suggest-
ed. The situation may be different for a few proteins [23]. In
such cases, using structures that structural studies have pre-
viously investigated might be appropriate [66]. Furthermore,
if the 3D structure of the protein hasn’t been determined yet
and is thus not available in the PDB, it should be built by
homology modeling [67].
Molecular docking needs the specification of some pa-
rameters. The PDB files often have deficient information and
therefore, they need to be corrected [68]. In the preparation
of the protein, hydrogens must be added, water should be
removed, charges must be assigned, and energy minimiza-
tion should be undertaken. There are several preparation
modules that fix common problems of PDB files [69].
Parameterization methods used vary depending on the
software [70]. AutoDock and SwissDock utilize an in-
program force field, whereas MOE uses AMBER and Le-
Dock uses CHARMM charges and atomic species. There-
fore, it is important to employ the same preparation protocol
in all docking procedures to compare the respective docking
results [71].
2.2.2. Preparation of Ligand
The structure of the ligands is drawn with programs like
ChemDraw [72] or is downloaded from chemical libraries or
databases like PubChem [73] and ZINC [74]. Before using
these structures in docking, energy minimization should be
undertaken [75].
It is recommended to visually examine the results of the
preparations of the target and ligand. Because some prepara-
tion methods can lead to mistakes in molecular descriptions,
such as incorrect connection, missing bonds, and abnormal
geometries. These errors often occur during the conversion of
one molecular format to another. Hence, it spreads easily [71].
After preparing the target and ligand, the binding site
should be determined and limited. It is possible to do this
step either by the specification of the coordinates manually
or by utilizing the coordinates of a ligand attached to the
protein. There are also programs that are used to calculate
the probable binding site [76]. The grid makes mapping the
binding area, which will be the center of docking calcula-
tions. The grid can be thought of as a box with known di-
mensions split into small squares in which the probe atoms
describe the contour of a possible interaction. Resolution and
size of the grid affect docking results [46].

4 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
Table 1. Molecular docking programs.
Program
Availability
Properties
AutoDock [45]
Free
Rigid body-flexible docking.
It is used with Autodock tools. Calculation of the grid maps is automatic.
AutoDock Vina [46]
Free
Rigid body-flexible docking.
It applies recurring local search global optimization. It is faster than AutoDock. It provides improved binding
affinity prediction with a new scoring function.
Dock [47]
Academic
Flexible docking.
It is widely applied to flexible targets and flexible ligands.
LeDock [48]
Academic
Flexible docking.
Since it gives results fastly with high accuracy, its use in virtual screening is recommended.
FlexX [49]
Commercial
Rigid body-flexible docking.
It can be utilized in virtual screening.
Glide [50]
Commercial
Ligands are flexible in this docking.
To decrease the software search range, it uses information about the area. It has XP (extra precision), SP (standard
precision), and highly efficient virtual screening modes.
GOLD [51]
Commercial
Flexible docking.
The evaluation of its accuracy and reliability appeared to give good results.
Plants [52]
Academic
It has a good balance between usage and efficiency. It allows calculating water exchange.
ICM [53]
Commercial
It gives the facility of both ligand-protein and protein-protein docking. It provides an ICM-Pro interface that
makes the docking process easy.
MOE [54]
Commercial
It has a good interface and intuitive aspect. It also consists of other tools that are used in protein and ligand preparation.
Surflex [55]
Commercial
For predocking minimization and post docking optimization, it uses procedures. It makes use of morphologic
similarity functions and fast pose production techniques.
LibDock [56]
Academic
LibDock depends on the matching of the polar and apolar binding site features of the target-ligand complex. As it is
driven by matching features rather than a molecular mechanics force field score, its performance attracts interest.
CDOCKER [57]
Free
CDOCKER (CHARMM based DOCKER) provides the advantages of full ligand flexibility, CHARMM force
field, and reasonable computation time. Flexible docking.
Fitted [58]
Free
Fitted can deal with both macromolecule flexibility and the presence of bridging water molecules.
Molegro [59]
Free
The program Molegro Virtual Docker (MVD) has four search algorithms and four native scoring functions. MVD
provides the opportunity of performing detailed statistical analysis of docking results when it is integrated with
other programs.
Fred/Hybrid [60]
Commercial
Fred uses the target structure solely to pose and score ligands. On the other hand, Hybrid uses both the target and
ligand structures to pose and score ligands. Hybrid has the ability to use multiple conformations of the target.
Fig. (1). Steps of molecular docking. (A higher resolution / colour version of this figure is available in the electronic copy of the article).

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 5
2.2.3. Determination of Docking Type
The choice of docking type to be used depends on the
needs of the researcher. If docking of several molecules at
the binding site of a protein at a specific pH, water, and sol-
ubility is desired, flexible docking programs may be pre-
ferred. However, if many more compounds (in thousands)
are to be scanned from databases, flexible docking methods
may be a bad option unless there is a high processor and a
fast computer. Therefore, the user can choose different dock-
ing methods according to the computer's capacity and the
target's properties [77].
2.2.4. Selection of the Best Docking Scoring Function
The best docking scoring function is selected depending
on the stability of the ligand-protein complex. It is difficult
to choose a suitable scoring function that gives a correct
binding pattern and the possible ligand. Theoretically, the
lower the binding free energy (ΔG) of a protein-ligand com-
plex, the more stable the complex is [78, 79].
Docking score is computed by various programs to iden-
tify and rank many poses of a ligand in a reasonable time
[80]. Scoring functions should be able to differentiate bind-
ers from nonbinders clearly. In addition to this, it should be
able to discriminate between correct and incorrect binding
modes of a ligand with high accuracy and in a reasonable
time [81]. Scoring functions are classified into three main
categories: Empirical, force field, and knowledge-based. In
empirical scoring functions, the free energy of binding is
calculated by adding hydrogen bonding, Van der Waals in-
teractions, electrostatics, hydrophobic interactions, and the
conformational free energy released when a ligand binds. In
the force field method, force field energy is computed using
molecular mechanics force fields similar to those used in
CHARMM and AMBER. This energy includes internal en-
ergies, coulombic interactions, including Van der Waals in-
teractions and hydrogen bonding. The entropy and solvent
energies are calculated separately. Knowledge-based scoring
functions are calculated by converting the frequencies of
ligand-protein atom interaction pairs into free energies using
Boltzmann distributions [15]. A single scoring function is
not perfect. Hence, it is possible to combine different scoring
functions to improve calculations with a single scoring func-
tion. This method is known as consensus scoring [80].
2.2.5. Docking Validation
Like any other technique, the docking process should
also be validated. The docking results are validated by re-
docking of reference ligands with targets and comparing the
RMSD (root mean square deviation) values, binding pose,
binding affinity, and coverage of the estimated bindings with
previously acquired results. If the ligand and target structures
are complex, it is recommended to carry out molecular dy-
namics studies. Molecular dynamics simulations can be uti-
lized to optimize the target before and after docking and to
provide flexibility, fix the complex after docking, calculate
the binding free energy including the solvent effect, and en-
sure the correct sequence of possible ligands [82]. At the end
of the process, the binding pose, binding residues and bind-
ing energies of the ligands are revealed (Fig. 2).
3. APPLICATIONS OF MOLECULAR DOCKING IN
DRUG DISCOVERY
With advances in docking algorithms, an increase in
open-access information on ligands and targets, the applica-
tions of molecular docking in drug discovery are rapidly
increasing [83]. In the early years, it was mainly used in the
investigation of the molecular interactions between ligands
and targets (15). However, these days the application scope
is wider and there is somewhat a shift in the application area.
Molecular docking has applications in virtual screening, tar-
get discovery and profiling, drug side effect prediction,
polypharmacology, and drug repurposing (Fig. 3) [24].
3.1. Virtual Screening
Virtual screening is used to find hits and lead compounds
from molecular databases according to scoring functions
[84]. The applications of docking in virtual screening have
increased with the combination of the method with other
Fig. (2). Binding residue points and binding pose of ciprofloxacin inside the binding site of DNA gyrase B. (A higher resolution / colour
version of this figure is available in the electronic copy of the article).

6 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
new applications. For example, the combination of molecu-
lar dynamics and free energy binding estimation methods
with docking has improved virtual screening [85].
Fig. (3). Applications of molecular docking in drug discovery. (A
higher resolution / colour version of this figure is available in the
electronic copy of the article).
These days there is a great effort worldwide to discover a
promising drug against the global pandemic, COVID-19.
There are efforts to discover drugs using SARS-CoV-2 (se-
vere acute respiratory syndrome coronavirus 2) targets such
as the structural spike (S) protein, envelope (E) protein,
membrane (M) protein, nucleocapsid (N) protein, and non-
structural proteins (Nsps) like the main protease (also called
3C-like protease (3CLpro, nsp5)), papain-like protease (PLpro,
nsp3), RNA-dependent RNA polymerase (RdRp, nsp12),
nsp15 endoribonuclease, nsp16 (2′-O-methyltransferase) and
nsp13 helicase. Host-based targets like angiotensin-
converting enzyme 2 (ACE2), transmembrane protease ser-
ine 2 (TMPRSS2), furin, and cathepsin are also used in this
effort [86]. Molecular docking has been used together with
other methods to support this effort [87]. For instance, re-
searchers performed an in silico screening of phytochemicals
and revealed that some of them could be effective against
SARS-CoV-2. Selected 154 herbal chemicals were docked to
five therapeutic protein targets of SARS-CoV-2 (proteases,
PLpro, SGp-RBD, RdRp, and ACE2) by using AutoDock
Vina. Using the docking score, the best 20 herbal chemicals
for each protein were screened for further investigation. By
using further computational analysis methods, 7 herbal
chemicals were proposed as potential SARS-CoV-2 inhibi-
tors for further in vitro and preclinical tests [88]. Similarly,
2000 molecules from the Selleck database of natural com-
pounds were screened by using ensemble docking against the
main protease (Mpro). The compounds that exhibited better
binding were filtered further by using Molecular Dynamics
(MD) simulations. Then, 11 natural compounds that were
found to bind to Mpro protease well were purchased and test-
ed in vitro. Finally, five promising Mpro protease inhibitor
natural compounds were determined [89].
In another similar work, a structural study was performed
to identify promising drug candidates to fight against
COVID-19. In this work, virtual screening together with
molecular docking was performed to look for potential inhib-
itors of the Mpro of SARS-CoV-2. Virtual screening was
done by using the Glide docking module. First, 50 molecules
from 2100 FDA-approved drugs in the ZINC database and
20 molecules from 400 natural products in the Spec database
were screened based on their docking score, glide energy,
and hydrogen bond interactions. Then, with XP glide dock-
ing and MD simulations, two compounds were suggested for
further experimental tests [90]. Similarly, hits from two in
silico screening studies were utilized in a wet-lab study to
identify potential Mpro inhibitors. The REAL Space or ZINC
databases were screening by ranking the molecules using
docking parameters [91, 92]. The promising compounds
were synthesized and assayed for their ability to inhibit the
activity of Mpro. Five compounds were found to inhibit the
enzyme in vitro [93].
3.2. Target Discovery and Profiling
Reverse docking allows the prediction of the biological
target of the respective molecule. As a result, it is a valuable
approach in computational target discovery and profiling
[94]. There are many docking approaches and algorithms for
reverse screening of a ligand against protein structure librar-
ies and evaluation of its binding affinities. However, the im-
plementation of these methods needs a convenient target
library [95]. There are several databases available for reverse
docking screening. PDTD (potential drug target database) is
a good example of familiar databases used in this area [96].
In addition to this, target libraries can be prepared manually
from databases such as PDB and TTD (therapeutic target
database) [97]. Reverse docking tools and web servers like
TarFisDock [98], idTarget [99], INVDOCK [100], Docko-
Matic [101] and SePreSA [102] are available for researchers.
In reverse docking screening, for a ligand, probable tar-
gets can be ranked by using the scoring functions used in the
programs [50]. For example, research using reverse docking
therapeutic mechanisms of astragaloside IV was investigat-
ed. In this study, all signaling pathways thought to be impli-
cated in the therapeutic actions of all cardiovascular disease
drugs approved by the FDA were considered. At the end of
the study, 39 putative targets were identified, and three of
them (CN, ACE, and JNK) were experimentally validated
[103].
3.3. Prediction of Drug Side Effects
Early detection of adverse drug effects is of great im-
portance in the drug discovery process. Drug candidates have
been reported to fail clinical trials mainly due to adverse
effects from unanticipated off-target interactions [104].
There are many computational approaches to support this
work. However, most model exercises require sufficient bio-
activity data or previously reported side effects [105]. To
predict side effects, molecular docking requires only struc-
tural information about the target. It is, therefore an im-
portant approach in predicting potential side effects of mole-
cules in the early phases without having detailed information
about the drug and bioactivity records. For example, re-

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 7
searchers conducted a reverse docking screening with torce-
trapib, a cholesteryl ester transfer protein inhibitor, to inves-
tigate the increase in mortality and cardiac events associated
with the side effects of the drug. Torcetrapib was docked
into a set of protein targets based on the enriched signaling
pathway. The results demonstrated that platelet-derived
growth factor receptor (PDGFR), hepatocyte growth factor
receptor (HGFR), IL-2 Receptor, and ErbB1 tyrosine kinase
might be the potential off-targets [106]. Databases that facili-
tate the identification of drug side effects have been devel-
oped. However, good performance in these predictions di-
rectly depends on the information in the databases. SIDER
(side effect resource) is one of the databases known in this
area [107]. Furthermore, by combining docking with ma-
chine learning (ML) and statistical approaches, advanced
screening methods, which also make drug side effect predic-
tions more advanced, were developed [108].
3.4. Polypharmacology
Polypharmacology expresses the identification of ligands
that interact with targets with a selected series of therapeutic
values. The pharmaceutical industry has concentrated on the
development of immensely selective drugs to avoid possible
side effects [109]. However, the high failure rate experienced
in the final phases of clinical tests as a result of lack of ther-
apeutic activity has led new drug designs to shift to
polypharmacology [110]. In this regard, molecular docking
provides a valuable opportunity as it permits the identifica-
tion of chemical structures that interact effectively with re-
lated targets simultaneously. It is difficult to design multitar-
get ligands for rational reasons. Furthermore, the choice of
protein structures to be utilized for docking can greatly in-
fluence the outcome of the design. This is particularly the
case when working with targets with remote binding sites
[111]. Docking is currently used in combination with other
in silico methods by considering the challenge of multitarget
drug design. Especially, several studies that comprise the
determination of multitarget ligands by applying docking
screening together with pharmacophore modeling have been
reported [112]. The determination of the first binary inhibitor
of Hsp90/B-Raf is an example. In this study, it has been
shown that substructure prefiltering and pharmacophore-led
docking can effectively be combined to look for polyphar-
macological ligands whose structures interact with different
targets. In another recent study, the potential of the cationic
pentapeptide Glu-Gln-Arg-Pro-Arg was assessed for its po-
tential role as an anticancer and anti-SARS-CoV-2. The
binding affinity of the peptide to integrins, Mpro, S protein,
and ACE2 was evaluated using molecular docking [113].
Polypharmacology workflows that combine docking with
other in silico methods were also followed [114].
There are docking-based web tools and platforms used to
investigate polypharmacology and determine the ligands'
multitarget activities. CANDO (computational analysis of
novel drug opportunities) [115] is an example for platforms
and DRAR-CPI (drug repositioning and adverse reactions
via chemical protein interactome) [116] is an example for
web servers.
3.5. Drug Repurposing
Drug repurposing is an established drug discovery way
that provides the opportunity to identify new therapeutic
applications for approved drugs, drug candidates under eval-
uation, natural products, or generally presynthesized ligands
[117]. Considering the wealth of information available in
public databases on ligands, targets, and diseases, efforts to
increase the application of the discovery strategies based on
in silico repurposing have increased over the last decades. In
silico repurposing methods have been shown to offer valua-
ble new opportunities in drug discovery and development
[118].
In this regard, molecular docking is among the most
widely used computational methods used for repurposing
ligands toward new therapeutic targets [119]. Docking lets
virtual screening of databases of approved drugs, phyto-
chemicals, or presynthesized compounds to the target of in-
terest in a reasonable time [95].
Recently, there are many studies that focus on repurpos-
ing existing drugs to combat COVID-19 [120]. For example,
researchers searched for commercially available drugs to
repurpose them against SARS-CoV-2 using in silico ap-
proaches. In this work, structure-based screening of ap-
proved drugs against Mpro and the serine protease TMPRSS2
of the novel coronavirus. Homology modeling was used to
generate the 3D structure of TMPRSS2. The structure-based
screening was performed by AutoDock Vina, and the result-
ing top-ranked hits were selected. With further molecular
docking using AutoDock 4.2, the best hits based on docking
score were screened. Then, with ADMET profile and drug-
likeness predictions, four approved drugs (talampicillin,
lurasidone, rubitecan, and loprazolam) from the drug library
were found to be potential inhibitors of Mpro and TMPRSS2
of the novel coronavirus. The stability of the complexes was
also checked by MD simulations [121].
In another study, researchers identified potential inhibi-
tors of Mpro of the novel coronavirus using in silico drug re-
purposing. Molecular docking calculations were carried out
using AutoDock 4.2 to select top-ranking drugs from the
DrugBank database. After the top-ranked approved drugs
from the database were filtered, 35 drugs with docking
scores of lower than -11.0 kcal/mol were picked for further
investigations. Then with MD simulations followed by MM-
GBSA (molecular mechanics–generalized Born surface area)
binding energy calculation, DB02388 and cobicistat
(DB09065) were found to be potential inhibitors for Mpro of
the novel coronavirus [122]. Similarly, mechanistic investi-
gation of the interaction of teicoplanin with MPro has been
done by molecular docking and molecular dynamics to re-
purpose it against SARS-CoV-2 [123]. Similarly, inhibitors
of its homolog, Hepatitis C Virus (HCV) protease, were in-
vestigated to repurpose them as Mpro inhibitors. 20 direct
acting antivirals of HCV were docked against Mpro and six of
them were found to be promising inhibitors [124]. In another
similar work, Remdesivir was found to be one of the hits for
Mpro inhibitors [125]. The available literature shows research
in discovering new molecules against novel coronavirus is
still in its infancy. Thus, virtual screening and drug repurpos-
ing of the available databases are the fastest options for the

8 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
discovery of potent drugs against the novel Coronavirus
[35].
Based on these promising results, it is possible to say that
docking is a valuable approach in drug repurposing. Espe-
cially, when it is combined with other computational ap-
proaches, such as ligand-based methods, its value increases
[24].
4. CURRENT STATUS OF MOLECULAR DOCKING
Molecular docking is broadly used in the academia and
pharmaceutical industry [11]. The wide scope of its applica-
tions, exemplified in the previous sections, demonstrates the
opportunity it provides for drug discovery. As is expected,
research works in molecular docking have been increasing.
Thus, the number of published articles in this area is rapidly
increasing (Fig. 4). To investigate the extensive usage of
molecular docking over the last two decades, the number of
documents available in publication databases has been
found. For this purpose, Scopus, PubMed, and ScienceDirect
search engines were used. These search engines were pre-
ferred since they were found to offer good search tools and
demonstrated satisfactory performance [126]. In each of
them, published documents were searched by using ‘molecu-
lar docking’ as a keyword. After extracting the number of
documents by year in the three engines, the average number
was calculated for each year starting from 2000. By using
these data, the graph of publications in the last two decades
was drawn (Fig. 4).
The results demonstrated that the number of publications
generated had nearly doubled every five years (Fig. 4). This
is in line with other similar studies conducted before. Cur-
rently, as illustrated by the publications, molecular docking
is widely used [20, 127].
Although many powerful programs are used in molecular
docking, there is no single program suitable for every sys-
tem. Consequently, users choose their preference depending
on the availability of the program, their needs, and their
computer capacity. They might also utilize more than one
program [21]. Therefore, molecular docking programs are
expected to have different popularity.
In this work, the relative popularity of selected docking
tools was also investigated. The total number of published
documents in Scopus, PubMed, and ScienceDirect search
engines was extracted. To find the total number of docu-
ments, ‘docking’ and the respective docking tools were used
as a keyword together. After the calculation of the average
total publication until 2020, the relative popularity graph was
drawn (Fig. 5). AutoDock was found to be the most popular
docking tool. Furthermore, GOLD and Glide were found to
be popular among commercial docking tools (Fig. 5). Previ-
ously reported studies also gave a similar popularity degree
[20, 127]. There has also been a considerable increase in the
popularity of AutoDock Vina in the last few years [128].
5. ADVANCES IN MOLECULAR DOCKING
The opportunities provided by molecular docking in the
drug discovery process are well known. However, intrinsic
factors limit the prediction performance of docking [129].
Thus, although it is essentially a stand-alone method in drug
design, these days it is used in combination with other com-
putational methods like ligand-based approaches, the rest of
structure-based approaches, quantum mechanics, machine
learning, and artificial intelligence (AI). This paves the way
to overcome some of the most important shortcomings of
docking [24].
5.1. Contribution of Ligant-Based Approaches
Ligand-based approaches have been used to identify ap-
propriate target structures for docking-based screening. The
ability of docking to differentiate active compounds from
inactive ones may have a high dependence on the 3D struc-
ture of the target used and the degree of similarity of the se-
lected ligands by screening against those compounds co-
crystallized in the target structure. Similarly, ligand-based
approaches have been utilized to increase the predictive po-
tential of docking screening [130]. For example, it can con-
Fig. (4). Publications in molecular docking. (A higher resolution / colour version of this figure is available in the electronic copy of the arti-
cle).

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 9
tribute to the evaluation of the 3D structure resemblance
between the binding pattern estimated by docking and the
experimentally detected binding pattern of the co-
crystallized ligand to the target structure. However, it
shouldn’t be forgotten that the possibility of using ligand-
based approaches in combination with docking applies only
to targets with a minimum of one reported co-crystallized
ligand [131].
5.2. Contribution of Structure-Based Approaches
Structure-based approaches, especially molecular dynam-
ics (MD) and binding free energy prediction, have been
broadly used in combination with docking to improve virtual
screening [132]. In this regard, MD is used to measure amino
acid flexibility in the binding site and to investigate greater
structural changes with potential accessibility to a given pro-
tein. Therefore, it is an efficient tool for the determination of
target structures for docking and evaluation of the stability of
the predicted complex [10]. The opportunities provided by
MD in silico screening, especially, address flexible targets
with few elucidated 3D structures. The contribution of bind-
ing free energy estimation to the improvement of virtual
screening has also been investigated. The output of currently
used docking algorithms might be affected by poor structural
sampling. They can also give incorrect binding energy pre-
dictions. Many approaches, such as BEAR (binding estima-
tion after refinement), MM-PBSA (molecular mechanics–
Poisson Boltzmann surface area), and MM-GBSA methods,
have been taken to address these issues [133]. These ap-
proaches have also been shown to improve virtual screening
and docking results [134].
5.3. Contribution of Quantum Mechanics
The contribution of Quantum Mechanics (QM) in im-
proving the prediction of binding free energy by molecular
docking is acknowledged [135]. The priority of molecular
mechanics (MM) scoring functions is speed rather than accu-
racy. Therefore, the reliability of predicting the free energy
of protein-ligand binding interactions is limited [136]. QM
calculations can be used to improve the prediction of binding
affinities, including re-scoring in docking [137]. The appli-
cation of QM calculations in docking rescoring brings a bet-
ter electrostatic interaction description and interaction ener-
gy. QM can also play its role in dealing with ionization and
tautomerism. Thus, QM-based scoring functions provide a
better correlation of calculated and experimental ligand af-
finities than the classical MM. This in turn, improves its role
in lead optimization [136].
5.4. Contribution of Machine Learning
Scoring and ranking candidate molecules by the calcula-
tion of binding affinity is a very challenging issue in molecu-
lar docking. Classical scoring functions need to simplify and
generalize several features of receptor-ligand interactions to
maintain efficiency, approachability, and accessibility [77].
In addition, classical scoring functions use linear regression
models, parametrically controlled learning methods that take
a predetermined functional form. Here, parametric methods
convert the input variables to the output forms with a prede-
fined function and adjust them in a theory-inspired manner
during the creation of the scoring function. This rigid scheme
often results in unadaptable scores that do not capture the
intrinsic nonlinearities of the data. Therefore, they show low
performance in cases that are not considered in their formu-
lations [138].
Machine learning algorithms can be used to improve or
replace predetermined function forms used in binding affini-
ty prediction in classical scoring. These have also been used
to identify binders/non-binders in virtual screening [139].
Machine learning, nonparametric learning, does not take the
form of predetermined functions. Instead, outputs are ex-
tracted from the input data. It can give a continuous output as
in nonlinear regression. This in turn allows for diverse and
accurate scoring. Random forest (RF)-Score is one of the
first machine learning scoring functions that outperform
classical scoring functions. In addition, logistic regression
and support vector machines (SVM) were used to improve
docking-based binding affinity predictions [138, 140, 141].
Fig. (5). The relative popularity of docking tools. (A higher resolution / colour version of this figure is available in the electronic copy of the
article).

10 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
5.5. Contribution of Artificial Intelligence
Artificial intelligence (AI) allows easy use of the ever-
growing open-access information sources in chemical, struc-
tural, and biological activity databases. This increases the
accuracy of binding affinity estimations [142]. In this con-
text, deep learning neural networks have been used in pose
generation and scoring [143]. The convolutional neural net-
work has been investigated in molecular docking by desig-
nating protein-ligand complexes as 3D cages. Deep learning
scoring functions have produced comparable and even supe-
rior results to machine learning and other non-neural network
algorithms [144, 145]. Machine learning might also be treated
as a member of this class. AI-based ML learns from the prop-
erties of the available data and then makes predictions on blind
data [146]. These approaches might not be preferable to newly
discovered therapeutic targets that have not been thoroughly
investigated yet and thus chemical, structural, and bioactivity
data about them are not available [24].
6. CHALLENGES IN MOLECULAR DOCKING
There are many difficulties in using docking tools and the
results of the study. It is reported that each program has its
limitations and flaws [147]. Therefore, programs cannot pro-
vide the same output with the same reliability. Furthermore,
the program may not perform well when the chemical struc-
ture processed exceeds the capacity of the developed soft-
ware. Therefore, it is important to continuously validate and
correct the developed software according to the new data.
[127]. Considering all these, not surprisingly, the acceptance
of the predictive tool results is still difficult. However, if the
current problems of the tools are addressed properly, the
value of the results and, therefore, the acceptance will in-
crease. In addition to this, if the resolution of the protein
structure available in the PDB or obtained from homology
modeling is poor, the docking result might not be reliable.
Thus, the selection of the structure of the protein to be used
in docking should be done with great care [148].
6.1. Accuracy of Docking
Docking methods are widely used to identify possible
ligands at the early stages of drug discovery and develop-
ment. There are many programs used to elucidate the interac-
tion of molecules with targets. Despite these programs, some
molecules have not yielded promising results when they are
tested in vivo [63]. Docking results may be interrogated due
to diverse issues. The first one is related to the use of protein
structure. Protein structures are generally available in com-
plexes with ligands in the PDB [149]. Researchers delete the
bound ligand to use the protein structure and do the docking
of the molecule being investigated. On the other hand, this
procedure may affect the docking output. The second im-
portant issue is the binding site environment. Drug candidate
molecules must bind to targets within the cell to exhibit their
activities. In some cases, even if the docking results exhibit
high binding in the in silico environment, they may give a
different result in the in vivo environment [63].
6.2. Properties of Ligand
It is impossible to predict the agonist or antagonist nature
of a ligand by docking. Docking studies give information
only about the binding mode and affinity of a molecule to-
wards a receptor [20]. To check the agonist or antagonist
properties of a molecule, experiments should be done in a
laboratory after the docking process. Therefore, it is recom-
mended not to overinterpret docking results regarding the
nature of the ligand unless other validations like lab experi-
ments, are performed [150].
Ligand preparation and conformation of ligands are also im-
portant in determining the docking results. In the ligand
preparation, molecules are ionized prior to docking. Howev-
er, the tautomeric state of the molecules is still a problem.
There is no clear way of using the variable tautomeric states
of the molecules to be docked [23].
6.3. Properties of Target
The quality of the structure of the target influences the
reliability of the molecular docking. Molecular structures
with the best geometrical parameters are chosen, but this
doesn’t guarantee that they are free of error. Thus, mecha-
nisms of filtering that will help in ensuring the quality of the
structures available in the databases, such as PDB are in
need [151].
In the preparation of the target, solvents and ligands in the
structure are usually removed. This leaves the binding pock-
et completely free. However, in the physiological state, the
environment is different. This leads to a discrepancy be-
tween the two conditions [152]. In recent years, there are
several attempts to use water molecules in the binding re-
gion. Nevertheless, there are still challenges in the way the
water is put around the binding site [153].
There are docking programs that use rigid protein struc-
tures. In real conditions, the target structure can fluctuate
depending on intrinsic and extrinsic factors, although it
spends more time in the lower energy states. Thus, docking
programs that keep the target rigid might give inaccurate
results [153]. Using programs that allow the target structure
to be flexible can be a solution here.
6.4. Search and Scoring Problems
Docking is difficult due to the various means of present-
ing two molecules in the 3D space together (three transla-
tional and three rotational degrees of freedom ). The search
algorithm implemented looks for all possible orientations
between two molecules by systemically translating and rotat-
ing one molecule over the other [154]. Many solutions can
be generated with a search algorithm. The solutions are
ranked according to their scores [155]. There are diverse
docking functions, and each program has its scoring system,
so there is no universal scoring function [156]. In general,
the correlation of docking scores with experimental binding
affinities is still poor. Each docking algorithm uses a scoring
function together with a search tool. Theoretically, the best
matching algorithms and scoring functions should be merged
to solve docking problems [157].
CONCLUSION
Molecular docking is a popular structure-based drug de-
sign method that predicts the interactions of small-molecule

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 11
ligands with the appropriate target. There are various power-
ful docking programs used for this purpose. Since no single
program is suitable for every system, choosing the most ap-
propriate one is recommended based on availability, need,
and computer capacity.
Molecular docking has many applications at various
stages of drug discovery. It has an established application,
especially in virtual screening and drug repurposing. Besides
the familiar diseases, there are several emerging diseases
nowadays. Thus, there is an urgent need for the discovery of
potent drugs against such diseases. Molecular docking is
playing an important role in the discovery of such drugs.
The challenges and limitations of molecular docking are
overcome by the involvement of other computational ap-
proaches. State-of-the-art computational methods like AI and
ML are expected to contribute much more in the near future.
Furthermore, with the increase in accessible biological and
chemical data, its application field is widening. As a result,
the use of molecular docking in the drug discovery process is
increasing. As a reflection of this, the number of publications
in this area has doubled almost every five years for the last
two decades.
The latest developments in other computational ap-
proaches had a substantial impact on molecular docking.
Therefore, there are advances in the quality of the generated
ligand-target binding modes and affinities. This will increase
the applicability of the resulting interactions. Thus, the role
of molecular docking in making the drug discovery process
rapid, economical, and more effective is expected to rise.
LIST OF ABBREVIATIONS
3CLpro = 3C-Like Protease
3D = Three Dimensional
ACE2 = Angiotensin-Converting Enzyme 2
AI = Artificial Intelligence
BEAR = Binding Estimation After Refinement
CADD = Computer-Aided Drug Design
CANDO = Computational Analysis of Novel Drug
Opportunities
COVID-19 = Corona Virus Disease 2019
CPU = Central Processing Unit
DRAR-CPI = Drug Repositioning and Adverse Reac-
tions via Chemical Protein Interactome
E = Envelope
GPU = Graphics Processing Unit
HCV = Hepatitis C Virus
HGFR = Hepatocyte Growth Factor Receptor
HIV = Human Immunodeficiency Virus
M = Membrane
MD = Molecular Dynamics
ML = Machine Learning
MM = Molecular Mechanics
MM-PBSA = Molecular Mechanics–Poisson Boltz-
mann Surface Area
Mpro = Main Protease
N = Nucleocapsid
Nsp = Nonstructural Proteins
PDB = Protein Data Bank
PDGFR = Platelet-Derived Growth Factor Receptor
PDTD = Potential Drug Target Database
PLpro = Papain-Like Protease
QM = Quantum Mechanics
QSAR = Quantitative Structure-Activity Relati-
onships
RdRp = RNA-dependent RNA Polymerase
RM = Random Forest
S = Spike
SARS-CoV-2 = Severe Acute Respiratory Syndrome Co-
ronavirus 2
SIDER = Side Effect Resource
SVM = Support Vector Machines
TMPRSS2 = TransMembrane Protease Serine 2
TTD = Therapeutic Target Database
CONSENT FOR PUBLICATION
Not applicable.
FUNDING
None.
CONFLICT OF INTEREST
The authors declare no conflict of interest, financial or
otherwise.
ACKNOWLEDGEMENTS
Declared none.
REFERENCES
[1] Prieto-Martínez, F.D.; López-López, E.; Eurídice Juárez-Mercado,
K.; Medina-Franco, J.L. Computational drug design methods-
Current and future perspectives. Silico Drug Des., 2019, 3(3), 19-
44.
http://dx.doi.org/10.1016/B978-0-12-816125-8.00002-X
[2] Kapetanovic, I.M. Computer-aided drug discovery and develop-
ment (CADDD): In silico-chemico-biological approach. Chem. Bi-
ol. Interact., 2008, 171(2), 165-176.
http://dx.doi.org/10.1016/j.cbi.2006.12.006 PMID: 17229415

12 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
[3] Barril, X. Computer-aided drug design: Time to play with novel
chemical matter. Expert Opin. Drug Discov., 2017, 12(10), 977-
980.
http://dx.doi.org/10.1080/17460441.2017.1362386 PMID:
28756685
[4] Deore, A.B.; Dhumane, J.R.; Wagh, R.; Sonawane, R. The stages
of drug discovery and development process. Asian J. Pharm. Res.
Dev., 2019, 7(6), 62-67.
http://dx.doi.org/10.22270/ajprd.v7i6.616
[5] Muhammed, M. T.; Aki-Yalcin, E. Pharmacophore modeling in
drug discovery: Methodology and current status. J. Turkish Chem.
Soc. Sect. A Chem., 2021, 8(3), 759-772.
[6] Surabhi, S.; Singh, B.K. Computer aided drug design : An over-
view. J. Drug Deliv. Ther., 2018, 8(5), 504-509.
http://dx.doi.org/10.22270/jddt.v8i5.1894
[7] Sliwoski, G.; Kothiwale, S.; Meiler, J.; Lowe, E.W., Jr Computa-
tional methods in drug discovery. Pharmacol. Rev., 2014, 66(1),
334-395.
http://dx.doi.org/10.1124/pr.112.007336 PMID: 24381236
[8] Ou-Yang, S.; Lu, J.; Kong, X.; Liang, Z.; Luo, C.; Jiang, H. Com-
putational drug discovery. Acta Pharmacol. Sin., 2012, 33(9),
1131-1140.
http://dx.doi.org/10.1038/aps.2012.109 PMID: 22922346
[9] Bisht, N.; Singh, B.K. Role of computer aided drug design in drug
development and drug discovery. Int. J. Pharm. Sci. Res., 2018,
9(4), 1405-1415.
http://dx.doi.org/10.13040/IJPSR.0975-8232.9(4).1405-15
[10] Salmaso, V.; Moro, S. Bridging molecular docking to molecular
dynamics in exploring ligand-protein recognition process: An
overview. Front. Pharmacol., 2018, 9, 923.
http://dx.doi.org/10.3389/fphar.2018.00923 PMID: 30186166
[11] Ferreira, L.; dos Santos, R.; Oliva, G.; Andricopulo, A. Molecular
docking and structure-based drug design strategies. Molecules,
2015, 20(7), 13384-13421.
http://dx.doi.org/10.3390/molecules200713384 PMID: 26205061
[12] Jones, L.H.; Bunnage, M.E. Applications of chemogenomic library
screening in drug discovery. Nat. Rev. Drug Discov., 2017, 16(4),
285-296.
http://dx.doi.org/10.1038/nrd.2016.244 PMID: 28104905
[13] Chahal, V.; Nirwan, S.; Kakkar, R. Combined approach of homol-
ogy modeling, molecular dynamics, and docking: Computer-aided
drug discovery. Physical Sci. Rev., 2019, 4(10), 1-15.
http://dx.doi.org/10.1515/psr-2019-0066
[14] Macalino, S.J.Y.; Billones, J.B.; Organo, V.G.; Carrillo, M.C.O. In
silico strategies in tuberculosis drug discovery. Molecules, 2020,
25(3), 665.
http://dx.doi.org/10.3390/molecules25030665 PMID: 32033144
[15] Hecht, D.; Fogel, G.B. Computational intelligence methods for
docking scores. Curr. Comput. Aided Drug Des., 2009, 5(1), 56-68.
http://dx.doi.org/10.2174/157340909787580863
[16] Meyers, J.; Fabian, B.; Brown, N. De novo molecular design and
generative models. Drug Discov. Today, 2021, 26(11), 2707-2715.
http://dx.doi.org/10.1016/j.drudis.2021.05.019 PMID: 34082136
[17] Mouchlis, V. D.; Afantitis, A.; Serra, A.; Fratello, M.; Papadiaman-
tis, A. G.; Aidinis, V.; Lynch, I.; Greco, D.; Melagraki, G. Advanc-
es in de novo drug design : From conventional to machine learning
methods. Int. J. Mol. Sci. 2021, 22(4), 1676.
[18] Sulimov, A.; Kutov, D.; Ilin, I.; Zheltkov, D.; Tyrtyshnikov, E.;
Sulimov, V. Supercomputer docking with a large number of de-
grees of freedom. SAR QSAR Environ. Res., 2019, 30(10), 733-749.
http://dx.doi.org/10.1080/1062936X.2019.1659412 PMID:
31547677
[19] Muhammed, M.T.; Kuyucuklu, G.; Kaynak-Onurdag, F.; Aki-
Yalcin, E. Synthesis, antimicrobial activity, and molecular model-
ing studies of some benzoxazole derivatives. Lett. Drug Des. Dis-
cov., 2022, 19(8), 757-768.
http://dx.doi.org/10.2174/1570180819666220408133643
[20] Chen, Y.C. Beware of docking! Trends Pharmacol. Sci., 2015,
36(2), 78-95.
http://dx.doi.org/10.1016/j.tips.2014.12.001 PMID: 25543280
[21] Tuccinardi, T.; Poli, G.; Romboli, V.; Giordano, A.; Martinelli, A.
Extensive consensus docking evaluation for ligand pose prediction
and virtual screening studies. J. Chem. Inf. Model., 2014, 54(10),
2980-2986.
http://dx.doi.org/10.1021/ci500424n PMID: 25211541
[22] Dar, A.M.; Mir, S. Molecular docking: Approaches, types, applica-
tions and basic challenges. J. Anal. Bioanal. Tech., 2017, 8(2), 8-
10.
http://dx.doi.org/10.4172/2155-9872.1000356
[23] Elokely, K.M.; Doerksen, R.J. Docking challenge: Protein sam-
pling and molecular docking performance. J. Chem. Inf. Model.,
2013, 53(8), 1934-1945.
http://dx.doi.org/10.1021/ci400040d PMID: 23530568
[24] Pinzi, L.; Rastelli, G. Molecular docking: Shifting paradigms in
drug discovery. Int. J. Mol. Sci., 2019, 20(18), 4331.
http://dx.doi.org/10.3390/ijms20184331 PMID: 31487867
[25] Phillips, M.A.; Stewart, M.A.; Woodling, D.L.; Xie, Z. Has molec-
ular docking ever brought. US Med., 2018, 1, 141-179.
http://dx.doi.org/10.5772/57353
[26] Ludovici, D.W.; De Corte, B.L.; Kukla, M.J.; Ye, H.; Ho, C.Y.;
Lichtenstein, M.A.; Kavash, R.W.; Andries, K.; de Béthune, M.P.;
Azijn, H.; Pauwels, R.; Lewi, P.J.; Heeres, J.; Koymans, L.M.H.;
de Jonge, M.R.; Van Aken, K.J.A.; Daeyaert, F.F.D.; Das, K.; Ar-
nold, E.; Janssen, P.A.J. Evolution of anti-HIV drug candidates.
Part 3: Diarylpyrimidine (DAPY) analogues. Bioorg. Med. Chem.
Lett., 2001, 11(17), 2235-2239.
http://dx.doi.org/10.1016/S0960-894X(01)00412-7 PMID:
11527705
[27] Janssen, P.A.J.; Lewi, P.J.; Arnold, E.; Daeyaert, F.; de Jonge, M.;
Heeres, J.; Koymans, L.; Vinkers, M.; Guillemont, J.; Pasquier, E.;
Kukla, M.; Ludovici, D.; Andries, K.; de Béthune, M.P.; Pauwels,
R.; Das, K.; Clark, A.D., Jr; Frenkel, Y.V.; Hughes, S.H.; Medaer,
B.; De Knaep, F.; Bohets, H.; De Clerck, F.; Lampo, A.; Williams,
P.; Stoffels, P. In search of a novel anti-HIV drug: Multidiscipli-
nary coordination in the discovery of 4-[[4-[[4-[(1E)-2-
cyanoethenyl]-2,6-dimethylphenyl]amino]-2- pyrimidi-
nyl]amino]benzonitrile (R278474, rilpivirine). J. Med. Chem.,
2005, 48(6), 1901-1909.
http://dx.doi.org/10.1021/jm040840e PMID: 15771434
[28] Zhang, P.; Bao, L.; Fan, J.; Jia, Z.J.; Sinha, U.; Wong, P.W.; Park,
G.; Hutchaleelaha, A.; Scarborough, R.M.; Zhu, B.Y.; Anthranila-
mide-Based, N. Anthranilamide-based N,N-dialkylbenzamidines as
potent and orally bioavailable factor Xa inhibitors: P4 SAR.
Bioorg. Med. Chem. Lett., 2009, 19(8), 2186-2189.
http://dx.doi.org/10.1016/j.bmcl.2009.02.114 PMID: 19297158
[29] Zhang, P.; Huang, W.; Wang, L.; Bao, L.; Jia, Z.J.; Bauer, S.M.;
Goldman, E.A.; Probst, G.D.; Song, Y.; Su, T.; Fan, J.; Wu, Y.; Li,
W.; Woolfrey, J.; Sinha, U.; Wong, P.W.; Edwards, S.T.; Arfsten,
A.E.; Clizbe, L.A.; Kanter, J.; Pandey, A.; Park, G.; Hutchaleelaha,
A.; Lambing, J.L.; Hollenbach, S.J.; Scarborough, R.M.; Zhu, B.Y.
Discovery of betrixaban (PRT054021), N-(5-chloropyridin-2-yl)-2-
(4-(N,N-dimethylcarbamimidoyl)benzamido)-5-
methoxybenzamide, a highly potent, selective, and orally effica-
cious factor Xa inhibitor. Bioorg. Med. Chem. Lett., 2009, 19(8),
2179-2185.
http://dx.doi.org/10.1016/j.bmcl.2009.02.111 PMID: 19297154
[30] von Itzstein, M.; Wu, W.Y.; Kok, G.B.; Pegg, M.S.; Dyason, J.C.;
Jin, B.; Van Phan, T.; Smythe, M.L.; White, H.F.; Oliver, S.W.;
Colman, P.M.; Varghese, J.N.; Ryan, D.M.; Woods, J.M.; Bethell,
R.C.; Hotham, V.J.; Cameron, J.M.; Penn, C.R. Rational design of
potent sialidase-based inhibitors of influenza virus replication. Na-
ture, 1993, 363(6428), 418-423.
http://dx.doi.org/10.1038/363418a0 PMID: 8502295
[31] Ellingson, S.R.; Miao, Y.; Baudry, J.; Smith, J.C. Multi-conformer
ensemble docking to difficult protein targets. J. Phys. Chem. B,
2015, 119(3), 1026-1034.
http://dx.doi.org/10.1021/jp506511p PMID: 25198248
[32] Sabe, V.T.; Ntombela, T.; Jhamba, L.A.; Maguire, G.E.M.; Goven-
der, T.; Naicker, T.; Kruger, H.G. Current trends in computer aided
drug design and a highlight of drugs discovered via computational
techniques: A review. Eur. J. Med. Chem., 2021, 224, 113705.
http://dx.doi.org/10.1016/j.ejmech.2021.113705 PMID: 34303871
[33] Chakraborty, R.; Parvez, S. COVID-19: An overview of the current
pharmacological interventions, vaccines, and clinical trials. Bio-
chem. Pharmacol., 2020, 180(July), 114184.
http://dx.doi.org/10.1016/j.bcp.2020.114184 PMID: 32739342
[34] Gurung, A.B.; Ali, M.A.; Lee, J.; Farah, M.A.; Al-Anazi, K.M. An
updated review of computer-aided drug design and its application
to COVID-19. BioMed Res. Int., 2021, 2021, 1-18.
http://dx.doi.org/10.1155/2021/8853056 PMID: 34258282

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 13
[35] Amin, S.A.; Jha, T. Fight against novel coronavirus: A perspective
of medicinal chemists. Eur. J. Med. Chem., 2020, 201(June),
112559.
http://dx.doi.org/10.1016/j.ejmech.2020.112559 PMID: 32563814
[36] A systematic review of RdRp of SARS-CoV-2 through artificial
intelligence and machine learning utilizing structure-based drug de-
sign strategy. Turk. J. Chem., 2021, 1-30.
http://dx.doi.org/10.3906/kim-2109-30
[37] Peele, K.A.; Potla Durthi, C.; Srihansa, T.; Krupanidhi, S.; Ayyaga-
ri, V.S.; Babu, D.J.; Indira, M.; Reddy, A.R.; Venkateswarulu, T.C.
Molecular docking and dynamic simulations for antiviral com-
pounds against SARS-CoV-2: A computational study. Inform.
Med. Unlocked, 2020, 19, 100345.
http://dx.doi.org/10.1016/j.imu.2020.100345 PMID: 32395606
[38] Serafim, M.S.M.; Gertrudes, J.C.; Costa, D.M.A.; Oliveira, P.R.;
Maltarollo, V.G.; Honorio, K.M. Knowing and combating the ene-
my: A brief review on SARS-CoV-2 and computational approaches
applied to the discovery of drug candidates. Biosci. Rep., 2021,
41(3), BSR20202616.
http://dx.doi.org/10.1042/BSR20202616 PMID: 33624754
[39] Meng, X.Y.; Zhang, H.X.; Mezei, M.; Cui, M. Molecular docking:
A powerful approach for structure-based drug discovery. Curr.
Comput. Aided Drug Des., 2011, 7(2), 146-157.
http://dx.doi.org/10.2174/157340911795677602 PMID: 21534921
[40] Prieto-Martínez, F.D.; Arciniega, M.; Medina-Franco, J.L. Molecu-
lar docking: Current advances and challenges. TIP Revi. Esp.
Cienc. Quim. Biol., 2018, 21(Suppl. 1), 1-23.
http://dx.doi.org/10.22201/fesz.23958723e.2018.0.143
[41] Lopes, P.E.M.; Guvench, O.; MacKerell, A.D., Jr Current status of
protein force fields for molecular dynamics simulations. Methods
Mol. Biol., 2015, 1215, 47-71.
http://dx.doi.org/10.1007/978-1-4939-1465-4_3 PMID: 25330958
[42] Hamzeh-mivehroud, M.; Sokouti, B.; Dastmalchi, S.; Islamia, J.M.;
Delhi, N.; Islamia, J.M.; Delhi, N.; Ambure, P.; Roy, K.; Anderluh,
M. The comparison of docking search algorithms and scoring func-
tions: An overview and case studies. In: Dastmalchi, S.; Hamzeh-
Mivehroud, M.; Babak, S.; Eds. Methods and Algorithms for Mo-
lecular Docking-Based Drug Design and Discovery. Hershey, PA:
IGI Global, 2016, pp. 99-127.
http://dx.doi.org/10.4018/978-1-5225-0115-2
[43] Wong, C.F. Flexible receptor docking for drug discovery. Expert
Opin. Drug Discov., 2015, 10(11), 1189-1200.
http://dx.doi.org/10.1517/17460441.2015.1078308 PMID:
26313123
[44] Pagadala, N.S.; Syed, K.; Tuszynski, J. Software for molecular
docking: A review. Biophys. Rev., 2017, 9(2), 91-102.
http://dx.doi.org/10.1007/s12551-016-0247-1 PMID: 28510083
[45] Morris, G.M.; Huey, R.; Lindstrom, W.; Sanner, M.F.; Belew,
R.K.; Goodsell, D.S.; Olson, A.J. AutoDock4 and AutoDock-
Tools4: Automated docking with selective receptor flexibility. J.
Comput. Chem., 2009, 30(16), 2785-2791.
http://dx.doi.org/10.1002/jcc.21256 PMID: 19399780
[46] Trott, O.; Olson, A.J. AutoDock Vina: Improving the speed and
accuracy of docking with a new scoring function, efficient optimi-
zation, and multithreading. J. Comput. Chem., 2010, 31(2), 455-
461.
http://dx.doi.org/10.1002/jcc.21334.AutoDock PMID: 19499576
[47] Allen, W.J.; Balius, T.E.; Mukherjee, S.; Brozell, S.R.; Moustakas,
D.T.; Lang, P.T.; Case, D.A.; Kuntz, I.D.; Rizzo, R.C. DOCK 6:
Impact of new features and current docking performance. J. Com-
put. Chem., 2015, 36(15), 1132-1156.
http://dx.doi.org/10.1002/jcc.23905 PMID: 25914306
[48] Unzue, A.; Xu, M.; Dong, J.; Wiedmer, L.; Spiliotopoulos, D.;
Caflisch, A.; Nevado, C. Fragment-based design of selective na-
nomolar ligands of the crebbp bromodomain. J. Med. Chem., 2016,
59(4), 1350-1356.
http://dx.doi.org/10.1021/acs.jmedchem.5b00172 PMID: 26043365
[49] Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. A fast flexible
docking method using an incremental construction algorithm. J.
Mol. Biol., 1996, 261(3), 470-489.
http://dx.doi.org/10.1006/jmbi.1996.0477 PMID: 8780787
[50] Friesner, R.A.; Banks, J.L.; Murphy, R.B.; Halgren, T.A.; Klicic,
J.J.; Mainz, D.T.; Repasky, M.P.; Knoll, E.H.; Shelley, M.; Perry,
J.K.; Shaw, D.E.; Francis, P.; Shenkin, P.S. Glide: A new approach
for rapid, accurate docking and scoring. 1. Method and assessment
of docking accuracy. J. Med. Chem., 2004, 47(7), 1739-1749.
http://dx.doi.org/10.1021/jm0306430 PMID: 15027865
[51] Verdonk, M.L.; Cole, J.C.; Hartshorn, M.J.; Murray, C.W.; Taylor,
R.D. Improved protein-ligand docking using GOLD. Proteins,
2003, 52(4), 609-623.
http://dx.doi.org/10.1002/prot.10465 PMID: 12910460
[52] Korb, O.; Stützle, T.; Exner, T.E. Empirical scoring functions for
advanced protein-ligand docking with PLANTS. J. Chem. Inf.
Model., 2009, 49(1), 84-96.
http://dx.doi.org/10.1021/ci800298z PMID: 19125657
[53] Abagyan, R.; Totrov, M.; Kuznetsov, D. ICM?A new method for
protein modeling and design: Applications to docking and structure
prediction from the distorted native conformation. J. Comput.
Chem., 1994, 15(5), 488-506.
http://dx.doi.org/10.1002/jcc.540150503
[54] Vilar, S.; Cozza, G.; Moro, S. Medicinal chemistry and the molecu-
lar operating environment (MOE): Application of QSAR and mo-
lecular docking to drug discovery. Curr. Top. Med. Chem., 2008,
8(18), 1555-1572.
http://dx.doi.org/10.2174/156802608786786624 PMID: 19075767
[55] Spitzer, R.; Jain, A.N. Surflex-Dock: Docking benchmarks and
real-world application. J. Comput. Aided Mol. Des., 2012, 26(6),
687-699.
http://dx.doi.org/10.1007/s10822-011-9533-y PMID: 22569590
[56] Rao, S.N.; Head, M.S.; Kulkarni, A.; LaLonde, J.M. Validation
studies of the site-directed docking program LibDock. J. Chem. Inf.
Model., 2007, 47(6), 2159-2171.
http://dx.doi.org/10.1021/ci6004299 PMID: 17985863
[57] Wu, G.; Robertson, D.H.; Brooks, C.L., III; Vieth, M. Detailed
analysis of grid-based molecular docking: A case study of
CDOCKER?A CHARMm-based MD docking algorithm. J. Com-
put. Chem., 2003, 24(13), 1549-1562.
http://dx.doi.org/10.1002/jcc.10306 PMID: 12925999
[58] Forli, S.; Huey, R.; Pique, M. E.; Sanner, M. F.; Goodsell, D. S.;
Olson, A. J. Computational protein–ligand docking and virtual drug
screening with the autodock suite. Nat. Protoc., 2016, 11(5), 905-
919.
http://dx.doi.org/10.1038/nprot.2016.051
[59] Bitencourt-Ferreira, G.; de Azevedo, W.F., Jr Molegro virtual
docker for docking. Methods Mol. Biol., 2019, 2053, 149-167.
http://dx.doi.org/10.1007/978-1-4939-9752-7_10 PMID: 31452104
[60] McGann, M. FRED and HYBRID docking performance on stand-
ardized datasets. J. Comput. Aided Mol. Des., 2012, 26(8), 897-
906.
http://dx.doi.org/10.1007/s10822-012-9584-8 PMID: 22669221
[61] Dong, D.; Xu, Z.; Zhong, W.; Peng, S. Parallelization of molecular
docking: A review. Curr. Top. Med. Chem., 2018, 18(12), 1015-
1028.
http://dx.doi.org/10.2174/1568026618666180821145215 PMID:
30129415
[62] Maia, E.H.B.; Medaglia, L.R.; da Silva, A.M.; Taranto, A.G. Mo-
lecular architect: A user-friendly workflow for virtual screening.
ACS Omega, 2020, 5(12), 6628-6640.
http://dx.doi.org/10.1021/acsomega.9b04403 PMID: 32258898
[63] Gupta, M.; Sharma, R.; Kumar, A. Docking techniques in pharma-
cology: How much promising? Comput. Biol. Chem., 2018,
76(June), 210-217.
http://dx.doi.org/10.1016/j.compbiolchem.2018.06.005 PMID:
30067954
[64] Ramírez, D.; Caballero, J. Is it reliable to take the molecular dock-
ing top scoring position as the best solution without considering
available structural data? Molecules, 2018, 23(5), 1038.
http://dx.doi.org/10.3390/molecules23051038 PMID: 29710787
[65] Muhammed, M.T.; Aki-Yalcin, E. Homology modeling in drug
discovery: Overview, current applications, and future perspectives.
Chem. Biol. Drug Des., 2019, 93(1), 12-20.
http://dx.doi.org/10.1111/cbdd.13388 PMID: 30187647
[66] Chi, P.B.; Liberles, D.A. Selection on protein structure, interaction,
and sequence. Protein Sci., 2016, 25(7), 1168-1178.
http://dx.doi.org/10.1002/pro.2886 PMID: 26808055
[67] Muhammed, M.T.; Son, Ç.D.; İzgü, F. Three dimensional structure
prediction of panomycocin, a novel Exo-β-1,3-glucanase isolated
from Wickerhamomyces anomalus NCYC 434 and the computa-
tional site-directed mutagenesis studies to enhance its thermal sta-

14 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
bility for therapeutic applications. Comput. Biol. Chem., 2019,
80(1), 270-277.
http://dx.doi.org/10.1016/j.compbiolchem.2019.04.006 PMID:
31054539
[68] Lohning, A.E.; Levonis, S.M.; Williams-Noonan, B.; Schweiker,
S.S. A practical guide to molecular docking and homology model-
ling for medicinal chemists. Curr. Top. Med. Chem., 2017, 17(18),
2023-2040.
http://dx.doi.org/10.2174/1568026617666170130110827 PMID:
28137238
[69] Warren, G.L.; Do, T.D.; Kelley, B.P.; Nicholls, A.; Warren, S.D.
Essential considerations for using protein–ligand structures in drug
discovery. Drug Discov. Today, 2012, 17(23-24), 1270-1281.
http://dx.doi.org/10.1016/j.drudis.2012.06.011 PMID: 22728777
[70] Voruganti, H.K.; Dasgupta, B. A novel volumetric criterion for
optimal shape matching of surfaces for protein-protein docking. J.
Comput. Des. Eng., 2018, 5(2), 180-190.
http://dx.doi.org/10.1016/j.jcde.2017.10.003
[71] Feher, M.; Williams, C.I. Numerical errors and chaotic behavior in
docking simulations. J. Chem. Inf. Model., 2012, 52(3), 724-738.
http://dx.doi.org/10.1021/ci200598m PMID: 22379951
[72] Cousins, K.R. Computer review of chemdraw ultra 12.0. J. Am.
Chem. Soc., 2011, 133(21), 8388.
http://dx.doi.org/10.1021/ja204075s PMID: 21561109
[73] Kim, S.; Chen, J.; Cheng, T.; Gindulyte, A.; He, J.; He, S.; Li, Q.;
Shoemaker, B.A.; Thiessen, P.A.; Yu, B.; Zaslavsky, L.; Zhang, J.;
Bolton, E.E. PubChem 2019 update: Improved access to chemical
data. Nucleic Acids Res., 2019, 47(D1), D1102-D1109.
http://dx.doi.org/10.1093/nar/gky1033 PMID: 30371825
[74] Sterling, T.; Irwin, J.J. ZINC 15-Ligand discovery for everyone. J.
Chem. Inf. Model., 2015, 55(11), 2324-2337.
http://dx.doi.org/10.1021/acs.jcim.5b00559 PMID: 26479676
[75] Andricopulo, A.; Guido, R.; Oliva, G. Virtual screening and its
integration with modern drug design technologies. Curr. Med.
Chem., 2008, 15(1), 37-46.
http://dx.doi.org/10.2174/092986708783330683 PMID: 18220761
[76] Feinstein, W.P.; Brylinski, M. Calculating an optimal box size for
ligand docking and virtual screening against experimental and pre-
dicted binding pockets. J. Cheminform., 2015, 7(1), 18.
http://dx.doi.org/10.1186/s13321-015-0067-5 PMID: 26082804
[77] Kitchen, D.B.; Decornez, H.; Furr, J.R.; Bajorath, J. Docking and
scoring in virtual screening for drug discovery: Methods and appli-
cations. Nat. Rev. Drug Discov., 2004, 3(11), 935-949.
http://dx.doi.org/10.1038/nrd1549 PMID: 15520816
[78] Cournia, Z.; Allen, B.; Sherman, W. Relative binding free energy
calculations in drug discovery: Recent advances and practical con-
siderations. J. Chem. Inf. Model., 2017, 57(12), 2911-2937.
http://dx.doi.org/10.1021/acs.jcim.7b00564 PMID: 29243483
[79] Luzhkov, V.B. Molecular modelling and free-energy calculations
of protein–ligand binding. Russ. Chem. Rev., 2017, 86(3), 211-230.
http://dx.doi.org/10.1070/RCR4610
[80] Kroemer, R.T. Structure-based drug design : Docking and scoring.
Curr. Protein Pept. Sci., 2007, 8(4), 312-328.
[81] Coupez, B.; Lewis, R.A. Docking and scoring-Theoretically easy,
Practically Impossible? Curr. Med. Chem., 2006, 13(25), 2995-
3003.
[82] Klepeis, J.L.; Lindorff-Larsen, K.; Dror, R.O.; Shaw, D.E. Long-
timescale molecular dynamics simulations of protein structure and
function. Curr. Opin. Struct. Biol., 2009, 19(2), 120-127.
http://dx.doi.org/10.1016/j.sbi.2009.03.004 PMID: 19361980
[83] Torres, P.H.M.; Sodero, A.C.R.; Jofily, P.; Silva-Jr, F.P. Key topics
in molecular docking for drug design. Int. J. Mol. Sci., 2019,
20(18), 4574.
http://dx.doi.org/10.3390/ijms20184574 PMID: 31540192
[84] Lionta, E.; Spyrou, G.; Vassilatis, D.; Cournia, Z. Structure-based
virtual screening for drug discovery: Principles, applications and
recent advances. Curr. Top. Med. Chem., 2014, 14(16), 1923-1938.
http://dx.doi.org/10.2174/1568026614666140929124445 PMID:
25262799
[85] Fan, J.; Fu, A.; Zhang, L. Progress in molecular docking. Quant.
Biol., 2019, 7(2), 83-89.
http://dx.doi.org/10.1007/s40484-019-0172-y
[86] Gil, C.; Ginex, T.; Maestro, I.; Nozal, V.; Barrado-Gil, L.; Cuesta-
Geijo, M.Á.; Urquiza, J.; Ramírez, D.; Alonso, C.; Campillo, N.E.;
Martinez, A. COVID-19: Drug targets and potential treatments. J.
Med. Chem., 2020, 63(21), 12359-12386.
http://dx.doi.org/10.1021/acs.jmedchem.0c00606 PMID: 32511912
[87] Saxena, A. Drug targets for COVID-19 therapeutics: Ongoing
global efforts. J. Biosci., 2020, 45(1), 87.
http://dx.doi.org/10.1007/s12038-020-00067-w PMID: 32661214
[88] Vardhan, S.; Sahoo, S.K. In silico ADMET and molecular docking
study on searching potential inhibitors from limonoids and triterpe-
noids for COVID-19. Comput. Biol. Med., 2020, 124, 103936.
http://dx.doi.org/10.1016/j.compbiomed.2020.103936 PMID:
32738628
[89] Rubio-Martínez, J.; Jiménez-Alesanco, A.; Ceballos-Laita, L.;
Ortega-Alarcón, D.; Vega, S.; Calvo, C.; Benítez, C.; Abian, O.;
Velázquez-Campoy, A.; Thomson, T.M.; Granadino-Roldán, J.M.;
Gómez-Gutiérrez, P.; Pérez, J.J. Discovery of diverse natural prod-
ucts as inhibitors of SARS-CoV-2 M pro protease through virtual
screening. J. Chem. Inf. Model., 2021, 61(12), 6094-6106.
http://dx.doi.org/10.1021/acs.jcim.1c00951 PMID: 34806382
[90] Sharma, P.; Vijayan, V.; Pant, P.; Sharma, M.; Vikram, N.; Kaur,
P.; Singh, T.P.; Sharma, S. Identification of potential drug candi-
dates to combat COVID-19: A structural study using the main pro-
tease (Mpro) of SARS-CoV-2. J. Biomol. Struct. Dyn., 2020, 0(0),
1-11.
http://dx.doi.org/10.1080/07391102.2020.1798286 PMID:
32741313
[91] Gorgulla, C.; Padmanabha Das, K.M.; Leigh, K.E.; Cespugli, M.;
Fischer, P.D.; Wang, Z.F.; Tesseyre, G.; Pandita, S.; Shnapir, A.;
Calderaio, A.; Gechev, M.; Rose, A.; Lewis, N.; Hutcheson, C.;
Yaffe, E.; Luxenburg, R.; Herce, H.D.; Durmaz, V.; Halazonetis,
T.D.; Fackeldey, K.; Patten, J.J.; Chuprina, A.; Dziuba, I.;
Plekhova, A.; Moroz, Y.; Radchenko, D.; Tarkhanova, O.;
Yavnyuk, I.; Gruber, C.; Yust, R.; Payne, D.; Näär, A.M.;
Namchuk, M.N.; Davey, R.A.; Wagner, G.; Kinney, J.; Arthanari,
H. A multi-pronged approach targeting SARS-CoV-2 proteins us-
ing ultra-large virtual screening. iScience, 2021, 24(2), 102021.
http://dx.doi.org/10.1016/j.isci.2020.102021 PMID: 33426509
[92] Ton, A.T.; Gentile, F.; Hsing, M.; Ban, F.; Cherkasov, A. Rapid
identification of potential inhibitors of SARS‐CoV‐2 main protease
by deep docking of 1.3 billion compounds. Mol. Inform., 2020,
39(8), 2000028.
http://dx.doi.org/10.1002/minf.202000028 PMID: 32162456
[93] Rossetti, G.G.; Ossorio, M.A.; Rempel, S.; Kratzel, A.; Dionellis,
V.S.; Barriot, S.; Tropia, L.; Gorgulla, C.; Arthanari, H.; Thiel, V.;
Mohr, P.; Gamboni, R.; Halazonetis, T.D. Non-covalent SARS-
CoV-2 Mpro inhibitors developed from in silico screen hits. Sci.
Rep., 2022, 12(1), 2505.
http://dx.doi.org/10.1038/s41598-022-06306-4 PMID: 35169179
[94] Huang, H.; Zhang, G.; Zhou, Y.; Lin, C.; Chen, S.; Lin, Y.; Mai,
S.; Huang, Z. Reverse screening methods to search for the protein
targets of chemopreventive compounds. Front Chem., 2018,
6(MAY), 138.
http://dx.doi.org/10.3389/fchem.2018.00138 PMID: 29868550
[95] Xu, X.; Huang, M.; Zou, X. Docking-based inverse virtual screen-
ing: Methods, applications, and challenges. Biophys. Rep., 2018,
4(1), 1-16.
http://dx.doi.org/10.1007/s41048-017-0045-8 PMID: 29577065
[96] Gao, Z.; Li, H.; Zhang, H.; Liu, X.; Kang, L.; Luo, X.; Zhu, W.;
Chen, K.; Wang, X.; Jiang, H. PDTD: A web-accessible protein da-
tabase for drug target identification. BMC Bioinformatics, 2008,
9(1), 104.
http://dx.doi.org/10.1186/1471-2105-9-104 PMID: 18282303
[97] Chen, X.; Ji, Z.L.; Chen, Y.Z. TTD: Therapeutic target database.
Nucleic Acids Res., 2002, 30(1), 412-415.
http://dx.doi.org/10.1093/nar/30.1.412 PMID: 11752352
[98] Li, H.; Gao, Z.; Kang, L.; Zhang, H.; Yang, K.; Yul, K.; Luo, X.;
Zhu, W.; Chen, K.; Shen, J. TarFisDock: A web server for identify-
ing drug targets with docking approach. Nucleic Acids Res., 2006,
34, 219-224.
http://dx.doi.org/10.1093/nar/gkl114
[99] Wang, J.C.; Chu, P.Y.; Chen, C.M.; Lin, J.H. idTarget: A web
server for identifying protein targets of small chemical molecules
with robust scoring functions and a divide-and-conquer docking
approach. Nucleic Acids Res., 2012, 40(W1), W393-W399.
http://dx.doi.org/10.1093/nar/gks496 PMID: 22649057
[100] Chen, Y.Z.; Zhi, D.G. Ligand-protein inverse docking and its po-
tential use in the computer search of protein targets of a small mol-
ecule. Proteins, 2001, 43(2), 217-226.

Molecular Docking: Principles, Advances, and Its Applications in Drug Discovery Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX 15
http://dx.doi.org/10.1002/1097-0134(20010501)43:2<217::AID-
PROT1032>3.0.CO;2-G PMID: 11276090
[101] Bullock, C.; Cornia, N.; Jacob, R.; Remm, A.; Peavey, T.; Weekes,
K.; Mallory, C.; Oxford, J.T.; McDougal, O.M.; Andersen, T.L.
DockoMatic 2.0: High throughput inverse virtual screening and
homology modeling. J. Chem. Inf. Model., 2013, 53(8), 2161-2170.
http://dx.doi.org/10.1021/ci400047w PMID: 23808933
[102] Yang, L.; Luo, H.; Chen, J.; Xing, Q.; He, L. SePreSA: A server
for the prediction of populations susceptible to serious adverse drug
reactions implementing the methodology of a chemical–protein in-
teractome. Nucleic Acids Res., 2009, 37 (Suppl. 2), W406-W412.
http://dx.doi.org/10.1093/nar/gkp312 PMID: 19417066
[103] Zhao, J.; Yang, P.; Li, F.; Tao, L.; Ding, H.; Rui, Y.; Cao, Z.;
Zhang, W. Therapeutic effects of astragaloside IV on myocardial
injuries: Multi-target identification and network analysis. PLoS
One, 2012, 7(9), e44938.
http://dx.doi.org/10.1371/journal.pone.0044938 PMID: 23028693
[104] Klein, E.; Bourdette, D. Postmarketing adverse drug reactions: A
duty to report? Neurol. Clin. Pract., 2013, 3(4), 288-294.
http://dx.doi.org/10.1212/CPJ.0b013e3182a1b9f0 PMID: 24195018
[105] Yoo, S.; Noh, K.; Shin, M.; Park, J.; Lee, K.H.; Nam, H.; Lee, D.
In silico profiling of systemic effects of drugs to predict unexpected
interactions. Sci. Rep., 2018, 8(1), 1612.
http://dx.doi.org/10.1038/s41598-018-19614-5 PMID: 29371651
[106] Fan, S.; Geng, Q.; Pan, Z.; Li, X.; Tie, L.; Pan, Y.; Li, X. Clarify-
ing off-target effects for torcetrapib using network pharmacology
and reverse docking approach. BMC Syst. Biol., 2012, 6(1), 152.
http://dx.doi.org/10.1186/1752-0509-6-152 PMID: 23228038
[107] Kuhn, M.; Letunic, I.; Jensen, L.J.; Bork, P. The SIDER database
of drugs and side effects. Nucleic Acids Res., 2016, 44(D1),
D1075-D1079.
http://dx.doi.org/10.1093/nar/gkv1075 PMID: 26481350
[108] Luo, H.; Fokoue-Nkoutche, A.; Singh, N.; Yang, L.; Hu, J.; Zhang,
P. Molecular docking for prediction and interpretation of adverse
drug reactions. Comb. Chem. High Throughput Screen., 2018,
21(5), 314-322.
http://dx.doi.org/10.2174/1386207321666180524110013 PMID:
29792141
[109] Ramsay, R.R.; Popovic-Nikolic, M.R.; Nikolic, K.; Uliassi, E.;
Bolognesi, M.L. A perspective on multi‐target drug discovery and
design for complex diseases. Clin. Transl. Med., 2018, 7(1), 3.
http://dx.doi.org/10.1186/s40169-017-0181-2 PMID: 29340951
[110] Anighoro, A.; Bajorath, J.; Rastelli, G. Polypharmacology: Chal-
lenges and opportunities in drug discovery. J. Med. Chem., 2014,
57(19), 7874-7887.
http://dx.doi.org/10.1021/jm5006463 PMID: 24946140
[111] Wei, D.; Jiang, X.; Zhou, L.; Chen, J.; Chen, Z.; He, C.; Yang, K.;
Liu, Y.; Pei, J.; Lai, L. Discovery of multitarget inhibitors by com-
bining molecular docking with common pharmacophore matching.
J. Med. Chem., 2008, 51(24), 7882-7888.
http://dx.doi.org/10.1021/jm8010096 PMID: 19090779
[112] Zhang, W.; Pei, J.; Lai, L. Computational multitarget drug design.
J. Chem. Inf. Model., 2017, 57(3), 403-412.
http://dx.doi.org/10.1021/acs.jcim.6b00491 PMID: 28166637
[113] Gasymov, O.K.; Celik, S.; Agaeva, G.; Akyuz, S.; Kecel-Gunduz,
S.; Qocayev, N.M.; Ozel, A.E.; Agaeva, U.; Bakhishova, M.; Ali-
yev, J.A. Evaluation of anti-cancer and anti-covid-19 properties of
cationic pentapeptide Glu-Gln-Arg-Pro-Arg, from rice bran protein
and its d-isomer analogs through molecular docking simulations. J.
Mol. Graph. Model., 2021, 108(April), 107999.
http://dx.doi.org/10.1016/j.jmgm.2021.107999 PMID: 34352727
[114] Anighoro, A.; Pinzi, L.; Marverti, G.; Bajorath, J.; Rastelli, G. Heat
shock protein 90 and serine/threonine kinase B-Raf inhibitors have
overlapping chemical space. RSC Advances, 2017, 7(49), 31069-
31074.
http://dx.doi.org/10.1039/C7RA05889F
[115] Chopra, G.; Samudrala, R. Exploring polypharmacology in drug
discovery and repurposing using the CANDO platform. Curr.
Pharm. Des., 2016, 22(21), 3109-3123.
http://dx.doi.org/10.2174/1381612822666160325121943 PMID:
27013226
[116] Luo, H.; Chen, J.; Shi, L.; Mikailov, M.; Zhu, H.; Wang, K.; He,
L.; Yang, L. DRAR-CPI: A server for identifying drug reposition-
ing potential and adverse drug reactions via the chemical–protein
interactome. Nucleic Acids Res., 2011, 39(Web Server is-
sue)(Suppl. 2), W492-W498.
http://dx.doi.org/10.1093/nar/gkr299 PMID: 21558322
[117] Keiser, M.J.; Setola, V.; Irwin, J.J.; Laggner, C.; Abbas, A.I.;
Hufeisen, S.J.; Jensen, N.H.; Kuijer, M.B.; Matos, R.C.; Tran, T.B.;
Whaley, R.; Glennon, R.A.; Hert, J.; Thomas, K.L.H.; Edwards,
D.D.; Shoichet, B.K.; Roth, B.L. Predicting new molecular targets
for known drugs. Nature, 2009, 462(7270), 175-181.
http://dx.doi.org/10.1038/nature08506 PMID: 19881490
[118] March-Vila, E.; Pinzi, L.; Sturm, N.; Tinivella, A.; Engkvist, O.;
Chen, H.; Rastelli, G. On the integration of in silico drug design
methods for drug repurposing. Front. Pharmacol., 2017, 8(MAY),
298.
http://dx.doi.org/10.3389/fphar.2017.00298 PMID: 28588497
[119] Kumar, S.; Kumar, S. Molecular Docking: A Structure-Based Ap-
proach for Drug Repurposing; Elsevier Inc: Amsterdam., 2019.
http://dx.doi.org/10.1016/B978-0-12-816125-8.00006-7
[120] Dotolo, S.; Marabotti, A.; Facchiano, A.; Tagliaferri, R. A review
on drug repurposing applicable to COVID-19. Brief. Bioinform.,
2021, 22(2), 726-741.
http://dx.doi.org/10.1093/bib/bbaa288 PMID: 33147623
[121] Elmezayen, A.D.; Al-Obaidi, A.; Şahin, A.T.; Yelekçi, K. Drug
repurposing for coronavirus (COVID-19): In silico screening of
known drugs against coronavirus 3CL hydrolase and protease en-
zymes. J. Biomol. Struct. Dyn., 2020, 39(8), 1-12.
http://dx.doi.org/10.1080/07391102.2020.1758791 PMID:
32306862
[122] Ibrahim, M.A.A.; Abdelrahman, A.H.M.; Hegazy, M.E.F. In-silico
drug repurposing and molecular dynamics puzzled out potential
SARS-CoV-2 Main Protease Inhibitors. J. Biomol. Struct. Dyn.,
2020, 39(15), 1-12.
http://dx.doi.org/10.1080/07391102.2020.1791958 PMID:
32684114
[123] Azam, F.; Eid, E.E.M.; Almutairi, A. Targeting SARS-CoV-2 main
protease by teicoplanin: A mechanistic insight by docking,
MM/GBSA and molecular dynamics simulation. J. Mol. Struct.,
2021, 1246, 131124.
http://dx.doi.org/10.1016/j.molstruc.2021.131124 PMID: 34305175
[124] Uddin, R.; Jalal, K.; Khan, K.; ul-Haq, Z. Re-purposing of hepatitis
C virus FDA approved direct acting antivirals as potential SARS-
CoV-2 protease inhibitors. J. Mol. Struct., 2022, 1250, 131920.
http://dx.doi.org/10.1016/j.molstruc.2021.131920 PMID: 34815586
[125] Hall, D.C., Jr; Ji, H.F. A search for medications to treat COVID-19
via in silico molecular docking models of the SARS-CoV-2 spike
glycoprotein and 3CL protease. Travel Med. Infect. Dis., 2020,
35(March), 101646.
http://dx.doi.org/10.1016/j.tmaid.2020.101646 PMID: 32294562
[126] Tober, M. PubMed, ScienceDirect, Scopus or Google Scholar –
Which is the best search engine for an effective literature research
in laser medicine? Med. Laser Appl., 2011, 26(3), 139-144.
http://dx.doi.org/10.1016/j.mla.2011.05.006
[127] Sousa, S.F.; Fernandes, P.A.; Ramos, M.J. Protein-ligand docking:
Current status and future challenges. Proteins, 2006, 65(1), 15-26.
http://dx.doi.org/10.1002/prot.21082 PMID: 16862531
[128] Vieira, T.F.; Sousa, S.F. Comparing autodock and vina in lig-
and/decoy discrimination for virtual screening. Appl. Sci. (Basel),
2019, 9(21), 4538.
http://dx.doi.org/10.3390/app9214538
[129] Chen, H.; Lyne, P.D.; Giordanetto, F.; Lovell, T.; Li, J. On evaluat-
ing molecular-docking methods for pose prediction and enrichment
factors. J. Chem. Inf. Model., 2006, 46(1), 401-415.
http://dx.doi.org/10.1021/ci0503255 PMID: 16426074
[130] Kumar, A.; Zhang, K.Y.J. Advances in the development of shape
similarity methods and their application in drug discovery. Front
Chem., 2018, 6(JUL), 315.
http://dx.doi.org/10.3389/fchem.2018.00315 PMID: 30090808
[131] Pinzi, L.; Caporuscio, F.; Rastelli, G. Selection of protein confor-
mations for structure-based polypharmacology studies. Drug Dis-
cov. Today, 2018, 23(11), 1889-1896.
http://dx.doi.org/10.1016/j.drudis.2018.08.007 PMID: 30099123
[132] Talevi, A.; Gavernet, L.; Bruno-Blanch, L. Combined virtual
screening strategies. Curr. Comput. Aided Drug Des., 2009, 5(1),
23-37.
http://dx.doi.org/10.2174/157340909787580854

16 Letters in Drug Design & Discovery, XXXX, Vol. XX, No. XX Muhammed and Aki-Yalcin
[133] Degliesposti, G.; Portioli, C.; Parenti, M.D.; Rastelli, G. BEAR, a
novel virtual screening methodology for drug discovery. SLAS Dis-
cov., 2011, 16(1), 129-133.
http://dx.doi.org/10.1177/1087057110388276 PMID: 21084717
[134] Guedes, I.A.; Pereira, F.S.S.; Dardenne, L.E. Empirical scoring
functions for structure-based virtual screening: Applications, criti-
cal aspects, and challenges. Front. Pharmacol., 2018, 9(Sep), 1089.
http://dx.doi.org/10.3389/fphar.2018.01089 PMID: 30319422
[135] Adeniyi, A.A.; Soliman, M.E.S. Implementing QM in docking
calculations: Is it a waste of computational time? Drug Discov. To-
day, 2017, 22(8), 1216-1223.
http://dx.doi.org/10.1016/j.drudis.2017.06.012 PMID: 28689054
[136] Caballero, J. The latest automated docking technologies for novel
drug discovery. Expert Opin. Drug Discov., 2020, 16(6), 1-21.
http://dx.doi.org/10.1080/17460441.2021.1858793 PMID:
33353444
[137] Ryde, U.; Söderhjelm, P. Ligand-binding affinity estimates sup-
ported by quantum-mechanical methods. Chem. Rev., 2016, 116(9),
5520-5566.
http://dx.doi.org/10.1021/acs.chemrev.5b00630 PMID: 27077817
[138] Ballester, P.J.; Mitchell, J.B.O. A machine learning approach to
predicting protein–ligand binding affinity with applications to mo-
lecular docking. Bioinformatics, 2010, 26(9), 1169-1175.
http://dx.doi.org/10.1093/bioinformatics/btq112 PMID: 20236947
[139] Ain, Q.U.; Aleksandrova, A.; Roessler, F.D.; Ballester, P.J. Ma-
chine-learning scoring functions to improve structure-based bind-
ing affinity prediction and virtual screening. Wiley Interdiscip. Rev.
Comput. Mol. Sci., 2015, 5(6), 405-424.
http://dx.doi.org/10.1002/wcms.1225 PMID: 27110292
[140] Korkmaz, S.; Zararsiz, G.; Goksuluk, D. MLViS: A web tool for
machine learning-based virtual screening in early-phase of drug
discovery and development. PLoS One, 2015, 10(4), e0124600.
http://dx.doi.org/10.1371/journal.pone.0124600 PMID: 25928885
[141] Chandak, T.; Mayginnes, J.P.; Mayes, H.; Wong, C.F. Using ma-
chine learning to improve ensemble docking for drug discovery.
Proteins, 2020, 88(10), 1263-1270.
http://dx.doi.org/10.1002/prot.25899 PMID: 32401384
[142] Yang, X.; Wang, Y.; Byrne, R.; Schneider, G.; Yang, S. Concepts
of artificial intelligence for computer-assisted drug discovery.
Chem. Rev., 2019, 119(18), 10520-10594.
http://dx.doi.org/10.1021/acs.chemrev.8b00728 PMID: 31294972
[143] Mogollon, D.C.; Fuentes, O.; Sirimulla, S. DLSCORE: A deep
learning model for predicting protein-ligand binding affinities.
ChemRxiv, 2018.
http://dx.doi.org/10.26434/chemrxiv.6159143.v1
[144] Jiménez, J.; Škalič, M.; Martínez-Rosell, G.; De Fabritiis, G. KDEEP:
Protein–ligand absolute binding affinity prediction via 3D-
convolutional neural networks. J. Chem. Inf. Model., 2018, 58(2),
287-296.
http://dx.doi.org/10.1021/acs.jcim.7b00650 PMID: 29309725
[145] Stokes, J.M.; Yang, K.; Swanson, K.; Jin, W.; Cubillos-Ruiz, A.;
Donghia, N.M.; MacNair, C.R.; French, S.; Carfrae, L.A.; Bloom-
Ackermann, Z.; Tran, V.M.; Chiappino-Pepe, A.; Badran, A.H.;
Andrews, I.W.; Chory, E.J.; Church, G.M.; Brown, E.D.; Jaakkola,
T.S.; Barzilay, R.; Collins, J.J. A deep learning approach to antibi-
otic discovery. Cell, 2020, 180(4), 688-702.e13.
http://dx.doi.org/10.1016/j.cell.2020.01.021 PMID: 32084340
[146] Jamal, S.; Khubaib, M.; Gangwar, R.; Grover, S.; Grover, A.;
Hasnain, S.E. Artificial Intelligence and Machine learning based
prediction of resistant and susceptible mutations in Mycobacterium
tuberculosis. Sci. Rep., 2020, 10(1), 5487.
http://dx.doi.org/10.1038/s41598-020-62368-2
[147] Huang, S.Y. Comprehensive assessment of flexible-ligand docking
algorithms: Current effectiveness and challenges. Brief. Bioinform.,
2018, 19(5), 982-994.
http://dx.doi.org/10.1093/bib/bbx030 PMID: 28334282
[148] Sarkar, A.; Sen, S. A comparative analysis of the molecular interac-
tion techniques for in silico drug design. Int. J. Pept. Res. Ther.,
2020, 26(1), 209-223.
http://dx.doi.org/10.1007/s10989-019-09830-6
[149] Rose, P.W.; Prlić, A.; Altunkaya, A.; Bi, C.; Bradley, A.R.; Chris-
tie, C.H.; Costanzo, L.D.; Duarte, J.M.; Dutta, S.; Feng, Z.; Green,
R.K.; Goodsell, D.S.; Hudson, B.; Kalro, T.; Lowe, R.; Peisach, E.;
Randle, C.; Rose, A.S.; Shao, C.; Tao, Y.P.; Valasatava, Y.; Voigt,
M.; Westbrook, J.D.; Woo, J.; Yang, H.; Young, J.Y.; Zardecki, C.;
Berman, H.M.; Burley, S.K. The RCSB protein data bank: Integra-
tive view of protein, gene and 3D structural information. Nucleic
Acids Res., 2017, 45(D1), D271-D281.
http://dx.doi.org/10.1093/nar/gkw1000 PMID: 27794042
[150] Prinz, F.; Schlange, T.; Asadullah, K. Believe it or not: How much
can we rely on published data on potential drug targets? Nat. Rev.
Drug Discov., 2011, 10(9), 712-713.
http://dx.doi.org/10.1038/nrd3439-c1 PMID: 21892149
[151] Markosian, C.; Di Costanzo, L.; Sekharan, M.; Shao, C.; Burley,
S.K.; Zardecki, C. Analysis of impact metrics for the Protein Data
Bank. Sci. Data, 2018, 5(1), 180212.
http://dx.doi.org/10.1038/sdata.2018.212 PMID: 30325351
[152] Wang, G.; Zhu, W. Molecular docking for drug discovery and
development: A widely used approach but far from perfect. Future
Med. Chem., 2016, 8(14), 1707-1710.
http://dx.doi.org/10.4155/fmc-2016-0143 PMID: 27578269
[153] Menchaca, T.M.; Juarez-Portilla, C.; Zepeda, R.C. Past, present,
and future of molecular docking. In: Gaitonde, V.; Karmakar, P.;
Trivedi, A. Drug Discovery and Development-New Advances;
London: IntechOpen, 2020, pp. 1-13.
[154] Ewing, T.J.A.; Kuntz, I.D. Critical evaluation of search algorithms
used in automated molecular docking. Comput. Appl. Biosci., 1997,
18, 1175-1189.
[155] Pyrkov, T.V.; Priestle, J.P.; Jacoby, E.; Efremov, R.G. Ligand-
specific scoring functions: Improved ranking of docking solutions.
SAR QSAR Environ. Res., 2008, 19(1-2), 91-99.
http://dx.doi.org/10.1080/10629360701844092 PMID: 18311637
[156] Yadava, U. Search algorithms and scoring methods in protein-
ligand docking. Endocrinol. Metabol. Inter. J., 2018, 6(6).
10.15406/emij.2018.06.00212
[157] Wang, R.; Lu, Y.; Wang, S. Comparative evaluation of 11 scoring
functions for molecular docking. J. Med. Chem., 2003, 46(12),
2287-2303.
http://dx.doi.org/10.1021/jm0203783 PMID: 12773034
DISCLAIMER: The above article has been published, as is, ahead-of-print, to provide early visibility but is not the final ver-
sion. Major publication processes like copyediting, proofing, typesetting and further review are still to be done and may lead to
changes in the final published version, if it is eventually published. All legal disclaimers that apply to the final published article
also apply to this ahead-of-print version.