Received: 9 December 2020 Revised: 27 January 2021 Accepted: 28 January 2021

DOI: 10.1002/wcms.1521

ADVANCED REVIEW

Gaussian accelerated molecular dynamics: Principles

and applications

Jinan Wang1 | Pablo R. Arantes2 | Apurba Bhattarai1 |

2 1 3
Rohaine V. Hsu | Shristi Pawnikar | Yu-ming M. Huang |
Giulia Palermo2,4 | Yinglong Miao1
1
Center for Computational Biology and Department of Molecular Biosciences, University of Kansas, Lawrence, Kansas, USA
2
Department of Bioengineering, University of California Riverside, Riverside, California, USA
3
Department of Physics & Astronomy, Wayne State University, Detroit, Michigan, USA
4
Department of Chemistry, University of California Riverside, Riverside, California, USA

Correspondence
Giulia Palermo, Department of
Abstract
Bioengineering and Department of Gaussian accelerated molecular dynamics (GaMD) is a robust computational
Chemistry, University of California method for simultaneous unconstrained enhanced sampling and free energy cal-
Riverside, 900 University Avenue,
Riverside, CA 92512. culations of biomolecules. It works by adding a harmonic boost potential to
Email: giulia.palermo@ucr.edu smooth biomolecular potential energy surface and reduce energy barriers.
GaMD greatly accelerates biomolecular simulations by orders of magnitude.
Yinglong Miao, Center for Computational
Biology and Department of Molecular Without the need to set predefined reaction coordinates or collective variables,
Biosciences, University of Kansas, 2030 GaMD provides unconstrained enhanced sampling and is advantageous for sim-
Becker Drive, Lawrence, KS 66047.
Email: miao@ku.edu
ulating complex biological processes. The GaMD boost potential exhibits a
Gaussian distribution, thereby allowing for energetic reweighting via cumulant
Funding information expansion to the second order (i.e., “Gaussian approximation”). This leads to
American Heart Association, Grant/
Award Number: 17SDG33370094;
accurate reconstruction of free energy landscapes of biomolecules. Hybrid
Extreme Science and Engineering schemes with other enhanced sampling methods, such as the replica-exchange
Discovery Environment (XSEDE), Grant/
GaMD (rex-GaMD) and replica-exchange umbrella sampling GaMD (GaREUS),
Award Numbers: TG-MCB160059, TG-
MCB180049; National Energy Research have also been introduced, further improving sampling and free energy calcula-
Scientific Computing Center (NERSC), tions. Recently, new “selective GaMD” algorithms including the Ligand GaMD
Grant/Award Number: M2874; National
(LiGaMD) and Peptide GaMD (Pep-GaMD) enabled microsecond simulations to
Institute of Health, Grant/Award
Numbers: R01EY027440, R01GM132572; capture repetitive dissociation and binding of small-molecule ligands and highly
National Science Foundation, Grant/ flexible peptides. The simulations then allowed highly efficient quantitative char-
Award Number: CHE-1905374
acterization of the ligand/peptide binding thermodynamics and kinetics. Taken
together, GaMD and its innovative variants are applicable to simulate a wide
Edited by: Peter R. Schreiner, Editor-in- variety of biomolecular dynamics, including protein folding, conformational
Chief
changes and allostery, ligand binding, peptide binding, protein–protein/nucleic
acid/carbohydrate interactions, and carbohydrate/nucleic acid interactions. In
this review, we present principles of the GaMD algorithms and recent applica-
tions in biomolecular simulations and drug design.

Jinan Wang and Pablo R. Arantes contributed equally to this study.

WIREs Comput Mol Sci. 2021;e1521. wires.wiley.com/compmolsci © 2021 Wiley Periodicals LLC. 1 of 32
https://doi.org/10.1002/wcms.1521
2 of 32 WANG ET AL.

This article is categorized under:

Structure and Mechanism > Computational Biochemistry and Biophysics
Molecular and Statistical Mechanics > Molecular Dynamics and Monte-
Carlo Methods
Molecular and Statistical Mechanics > Free Energy Methods

KEYWORDS
drug binding, free energy calculations, enhanced sampling, membrane proteins, protein/nucleic
acid complexes

1 | INTRODUCTION

Biological processes are mediated by biomolecules such as proteins, nucleic acids, lipids, and carbohydrates. Biomolecules
often visit different functional conformations during various biological functions, including cellular signaling, protein fold-
ing, gene translation/editing, and biomolecular recognition.1–4 The underlying free energy landscapes of biomolecules
determine their conformations.5,6 Molecular dynamics (MD) is an advanced technique that allows us to simulate biomo-
lecular dynamics at an atomistic level.7 It is now possible to run longer and cheaper MD simulations with remarkable
advances in computing hardware (e.g., the Anton supercomputer and GPUs) and software developments.8 Even so, con-
ventional MD (cMD) is often limited to typically hundreds of nanoseconds to tens of microseconds.9–12 On the other hand,
many biological processes of interest take place over milliseconds or even longer timescales, due to high energy barriers
(e.g., 8–12 kcal/mol).1,13,14 Due to this gap, it remains challenging to sufficiently sample different conformations and accu-
rately calculate free energy profiles of biomolecules through cMD simulations.
To overcome the above challenges, numerous enhanced sampling techniques have been introduced since the dawn
of MD as reviewed in a number of previous articles.15–18 One class of these methods use predefined collective variables
(CVs) or reaction coordinates, including umbrella sampling (US),19,20 metadynamics,21,22 adaptive biasing force
(ABF),23,24 steered MD (SMD),25 conformational flooding,26,27 and so on. Typical CVs include root-mean-square devia-
tion (RMSD) relative to a reference conformation, dihedrals, atom distances, eigenvectors of principal component anal-
ysis (PCA),27 and so forth. These methods greatly improve the sampling of biomolecular dynamics and the accuracy of
free energy calculations along with the chosen CVs. However, it is rather challenging to define proper CVs in prior
because the system needs to be studied in detail beforehand. Furthermore, the predefined CVs could largely limit sam-
pling of the conformational space during the biasing simulations. This usually slows convergence of the simulations
and suffers from the “hidden energy barrier” problem once crucial CVs are missing in the simulation setup.22
Another kind of enhanced sampling techniques have been introduced without using predefined CVs, including replica
exchange molecular dynamics (REMD)28,29 or parallel tempering,30 self-guided Langevin or molecular dynamics,–33 essen-
tial energy space random walk,34–36 and accelerated molecular dynamics (aMD).37 In particular, Voter introduced aMD
by adding a boost potential in non-barrier regions to accelerate infrequent transitions in solids.38 Hamelberg et al. further
developed this technique to perform biomolecular simulations.37 The boost potential in aMD enables simulations to sam-
ple different low-energy conformational states by smoothing the system potential energy surface and reducing the energy
barriers.37,39 Despite the advantage of unconstrained enhanced sampling, aMD can suffer from high statistical noise,
affecting the description of the correct statistical ensemble.40 In detail, the canonical ensemble average is reached by
reweighting each point in the configuration space on the modified potential energy surface by the strength of the
Boltzmann factor of the bias energy, at that particular point. Using the early aMD method, this has shown to lead to high
statistical noise, since the points with the largest biases dominate the reweighted result.41–44 In comparison with the CV-
biasing methods, aMD has typically much higher boost potential with wider distributions (tens to hundreds of kcal/
mol),42 making it very challenging to accurately reweight free energies from aMD simulations, especially for biological
macromolecules.43,45,46 This issue can be severe for large biomolecular systems, such as transmembrane proteins and ribo-
nucleoproteins, where the standard reweighting procedure has often been prohibitive, given the large statistical noise.42
Gaussian accelerated molecular dynamics (GaMD) has been developed to smooth the surface of potential energy
with a harmonic boost potential, following three newly formulated enhanced sampling principles.47 Similar to the pre-
vious aMD, no predefined CV is needed for GaMD simulations. Furthermore, the new harmonic boost potential in
WANG ET AL. 3 of 32

GaMD exhibits a Gaussian distribution, which enables us to accurately recover the original biomolecular free energy
landscapes by Gaussian approximation, that is, cumulant expansion to the second order. This useful scheme substan-
tially reduces the statistical noise, thereby overcoming the limitations of the early aMD methodology (vide supra).
Therefore, GaMD simultaneously enables enhanced sampling without any constraints and accurately calculates free
energy landscapes of biomolecules. As previously reported,48,49 GaMD has been successfully applied to simulate ligand
binding,47,50,51 protein folding,47,51 activation of G-protein-coupled receptors (GPCRs),50 human dystonia related pro-
tein,52 ion channels,53 agonists and antagonist binding in the μ-OR,54,55 virus enzymes,56,57 bacterial effector proteins,58
and so forth.
In addition, GaMD has been combined with REMD to further improve conformational sampling and free energy
calculations.59,60 More recently developed “selective GaMD” algorithms, including Ligand GaMD (LiGaMD)61 and Pep-
tide GaMD (Pep-GaMD),62 have enabled unprecedented microsecond simulations to capture repetitive binding and dis-
sociation of small-molecule ligands and highly flexible peptides. Accurate ligand/peptide binding free energies and
kinetic rate constants are thus calculated through the selective GaMD simulations.
In this review, we will present the principles and the most recent applications of GaMD. Robust GaMD has been
established for advanced simulation studies of a wide range of biomolecular systems, especially the protein–nucleic acid
interactions63–65 such as the CRISPR (clustered regularly interspaced short palindromic repeats)–Cas9 gene-editing
system,66,67 protein–protein/peptide interactions,68–72 protein-ligand binding,55,56,73–76 protein folding,77 protein
enzymes,52,58,78–92 membrane proteins (including GPCRs,68,69,73,76,93–95 ion channels,53,96 and γ-secretase97) and carbo-
hydrates,98–100 as well as drug design.101,102

2 | THEORY

2.1 | Gaussian accelerated molecular dynamics

A harmonic boost potential is added in GaMD to smooth the system potential energy surface andn enhanceothe confor-
* *
mational samplingof biomolecules (Figure 1).47 Consider a system with N atoms at positions
r = r 1 , … r N , when the
*
system potential V r is lower than a threshold energy E, a boost potential is added as:

  1   2  
* * *
ΔV r = k E −V r , V r < E, ð1Þ
2
     
* * *
where k is the harmonic force constant. The modified system potential, V  r = V r + ΔV r is given by:

    1   2  
* * * *
V  r = V r + k E −V r ,V r < E : ð2Þ
2

F I G U R E 1 Scheme illustration of Gaussian accelerated

molecular dynamics (GaMD). When the threshold energy is set to
the maximum potential (E = Vmax), the system potential energy
surface is smoothened by adding a harmonic boost potential that
follows Gaussian distribution. The coefficient k0 in the range of 0–1
determines the magnitude of the applied boost potential. With
greater k0, higher boost potential is added to the original energy
surface in conventional molecular dynamics (cMD), which provides
enhanced sampling of biomolecules across decreased energy
barriers. Adapted with permission from Miao et al. (2015).
Copyright 2015 American Chemical Society. https://pubs.acs.org/
doi/abs/10.1021/acs.jctc.5b00436. Further permissions related to the
material excerpted should be directed to the American Chemical
Society
4 of 32 WANG ET AL.

 
*
Otherwise, when the system potential is above the threshold energy, that is, V r ≥E, the boost potential is set to zero
   
* *
and V  r = V r .

Three enhanced sampling principles are applied to the boost potential in GaMD to smooth the potential energy
   
* *
surface. First, for any two arbitrary potential values V 1 r and V 2 r found on the original energy surface, if
   
* *
V 1 r < V 2 r , ΔV should be a monotonic function that does not change the relative order of the biased potential
   
* *
values, that is, V 1  r < V 2  r . By replacing V  ð
r Þ with Equation (2) and isolating E, we then obtain:

1 h *   i 1
*
E< V1 r + V2 r +  ð3Þ
2 k

   
* *
Second, if V 1 r < V 2 r , the potential difference observed on the smoothened energy surface should be smaller than that of
         
* * * * *
the original, that is, V 2 r −V 1 r < V 2 r − V 1 r . Similarly, by replacing V  r with Equation (2), we can derive:

1 h *  i
*
E> V1 r + V2 r ð4Þ
2
   
* *
With V min ≤V 1 r < V 2 r ≤V max , we need to set the threshold energy E in the following range by combining Equa-
tions (3) and (4):

1
V max ≤ E ≤ V max + , ð5Þ
k

where Vmin and Vmax are the system minimum and maximum potential energies. To ensure that Equation (5) is valid,
1
V max ≤ V min + and k has to satisfy:
k

1
k≤  ð6Þ
V max −V min

Let us define k  k 0 × V max −1 V min , then 0 < k0 ≤ 1. As illustrated in Figure 1, k0 determines the magnitude of the applied
boost potential. With higher k0, larger boost potential is added to the potential energy surface, which facilitates
enhanced sampling of biomolecules across decreased energy barriers.

Third, in order to ensure accurate reweighting using cumulant expansion to the second order,42 the standard devia-
tion of ΔV needs to be small enough (i.e., narrow distribution):
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s 2
∂ΔV  
σ ΔV = V = V avg σ 2V = k E −V avg σ V ≤ σ 0 , ð7Þ
∂V

where Vavg and σ V are the average and standard deviation of the system potential energies, σ ΔV is the standard deviation
of ΔV with σ 0 as a user-specified upper limit (e.g., 10kBT) for accurate reweighting.
Provided Equation (5) that gives the range of threshold energy E, when E is set to the lower bound E = Vmax, we
substitute in E and k, and obtain:

σ 0 V max − V min
k0 ≤ × : ð8Þ
σ V V max −V avg
WANG ET AL. 5 of 32

Let us define the right-hand side in Equation (8) as k 00 = σσV0 × VVmax − V min
max − V avg
. For efficient enhanced sampling with the highest
possible acceleration, k0 can then be set to its upper bound as:
 
 σ 0 V max −V min
k 0 = min 1:0, k 00 = min 1:0, ×  ð9Þ
σ V V max − V avg

The larger σ ΔV is obtained from the original potential energy surface (particularly for large biomolecules), the smaller
k0 may be applicable to allow for accurate reweighting. Alternatively, when the threshold energy E is set to its upper
bound E = V min + k1 according to Equation (5), we substitute in E and k in Equation (7) and obtain:
 
σ0 V max −V min
k0 ≥
1− × : ð10Þ
σV V avg −V min
 
Let us define the right-hand side in Equation (10) as k000  1 − σσV0 × VVmax − V min
avg − V min
. Note that a smaller k0 will give higher
threshold energy E, but smaller force constant k. When 0 < k0 ≤1, k0 can be set to either k000 for the highest threshold
00

energy E or its upper bound 1.0 for the greatest force constant k. In this regard, k 0 = k 000 is applied in the current GaMD.
Otherwise, k0 is calculated using Equation (9).
Given E and k0, we can calculate the boost potential as:

1 1
ΔV ð
r Þ = k 0 ðE −V ð
r ÞÞ2 , V ð
r Þ < E: ð11Þ
2 V max −V min

GaMD provides different options to add only the total potential boost ΔVp, only dihedral potential boost ΔVD, or the
dual potential boost (both ΔVp and ΔVD). The dual-boost GaMD generally provides higher acceleration than the other
two types of simulations for enhanced sampling.39 The simulation parameters comprise of the threshold energy values
and the effective harmonic force constants, k0p and k0D for the total and dihedral potential boost, respectively.

2.2 | Energetic reweighting of GaMD for free energy calculations

For simulations of a biomolecular system, the probability distribution along a selected reaction coordinate A(r) is writ-
ten as p*(A), where r denotes the atomic positions r = {r1, …, rN}. Given the boost potential ΔV(r) of each frame, p*(A)
can be reweighted to recover the canonical ensemble distribution, p(A), as:

  eβΔV ðrÞ
p Aj = p Aj
j
, j = 1, …, M, ð12Þ
P
M
hp  ðA i ÞeβΔV ðrÞ i
i=1 i

where M is the number of bins, β = kBT and heβΔV(r)ij is the ensemble-averaged Boltzmann factor of ΔV(r) for simulation
frames found in the jth bin. To reduce the energetic noise, the ensemble-averaged reweighting factor can be approxi-
mated using cumulant expansion103,104:
( )
X
∞ k
β
βΔV
e = exp Ck , ð13Þ
k=1
k!

where the first three cumulants are given by:

C 1 = hΔV i,
C 2 = ΔV 2 − hΔV i2 = σ 2ΔV , ð14Þ
2
C 3 = ΔV 3
−3hΔV i hΔV i + 2 ΔV : 3
6 of 32 WANG ET AL.

When the boost potential follows near-Gaussian distribution, cumulant expansion to the second order (or “Gaussian
Approximation”) provides the accurate approximation for free energy calculations.42 The reweighted free energy
F(A) = − kBT ln p(A) is calculated as:

1X 2
βk
F ðAÞ = F  ðAÞ − Ck + F c , ð15Þ
β k = 1 k!

where F*(A) = − kBT ln p*(A) is the modified free energy obtained from GaMD simulation and Fc is a constant.
To characterize the extent to which ΔV follows a Gaussian distribution, its distribution anharmonicity γ is calcu-
lated as42:
ð∞
1 
γ = Smax −SΔV = ln 2πeσ 2ΔV + pðΔV ÞlnðpðΔV ÞÞdΔV , ð16Þ
2 0

where ΔV is dimensionless as divided by kBT with kB and T being the Boltzmann constant and system temperature,
respectively, and Smax = 12 ln 2πeσ 2ΔV is the maximum entropy of ΔV.42 When γ is zero, ΔV follows exact Gaussian dis-
tribution with sufficient sampling. Reweighting by approximating the exponential average term with cumulant
expansion to the second order is able to accurately recover the original free energy landscape. As γ increases, the
ΔV distribution becomes less harmonic and the reweighted free energy profile obtained from cumulant expansion
to the second order would deviate from the original. The anharmonicity of ΔV distribution serves as an indicator
of the enhanced sampling convergence and accuracy of the reweighted free energy. Nevertheless, with the GaMD
theoretical framework, the GaMD boost potential does not change shape of the biomolecular overall energy land-
scape. A near Gaussian distribution is achieved for the GaMD boost potential. A toolkit of Python scripts for
GaMD/aMD reweighting “PyReweighting”42 is developed and distributed free of charge at http://miao.compbio.
ku.edu/PyReweighting/.

2.3 | Replica exchange–GaMD

Replica exchange and GaMD have been combined in a rex-GaMD approach to further improve the sampling and free
energy calculations of biomolecules.59 According to Equation (11), both the threshold energy E and the effective force
constant k0 could adjust the boost potential. Therefore, two versions of rex-GaMD were proposed: force constant rex-
GaMD and threshold energy rex-GaMD. During simulations of force constant rex-GaMD, the boost potential can be
exchanged between replicas, in which the threshold energy is fixed and harmonic force constants are different. Whereas
the algorithm of threshold energy rex-GaMD tends to switch the threshold energy between lower and upper bounds for
generating different levels of boost potential.
The rex-GaMD simulations allow replicas exchanged between each pair of neighboring σ 0P or threshold energy
based on the probability that meets the Metropolis criterion. In the simulation system, each state x can be weighted by
the Boltzmann factor,
 
1
W B ðx Þ = exp H ðx Þ , ð17Þ
kB T

where kB is the Boltzmann constant, T is the system temperature and H(x) is the system Hamiltonian. The weight factor
for the state X here is given by the product of the Boltzmann factor of each replica:
 X 
N 1
W RE ðX Þ = exp − i=1 k T
H ðx i Þ ð18Þ
B

where N is the number of total states. Thus, the replica-exchange probability can be written as w(Xi ! Xj), which needs
to meet the Metropolis criterion to calculate the exchange probability:
WANG ET AL. 7 of 32

 
w X i ! X j = min 1:0,eΔ , ð19Þ

 
where Xi and Xj are the states of the two nearby replicas, and Δ = kB1T V i −V j and V*i and V*j are the total modified
system potential energies calculated from the last confirmation of the GaMD simulations at replica i and j. These
exchange processes will keep repeating until the end of the simulation. The rex-GaMD simulations were tested on three
model systems, including the alanine dipeptide, chignolin, and HIV protease, demonstrating that the distribution width
of the boost potential is narrowed down, and the system conformational space is enhanced sampled.
Recently, Sugita et al.60 proposed another approach (GaREUS) that combined GaMD with replica-exchange
umbrella sampling (REUS). GaREUS was successfully demonstrated on accurate calculations of free energy landscapes
underlying the N-glycan equilibration, conformational change of adenylate kinase, and chignolin folding. The computa-
tional resource for GaREUS was the same as that required for REUS, while the sampling in GaREUS was more efficient
than REUS or GaMD.

2.4 | Ligand Gaussian accelerated molecular dynamics

Based on GaMD, LiGaMD61 has been proposed to more efficiently simulate both binding and dissociation of small-
molecule ligands for calculating the ligand binding free energies and kinetics. For such simulations, the system contains
n
ligand o
L, protein n
P and the biological o
environment E. The system comprises N atoms with their coordinates r 
* * * *
r 1 , …, r N and momenta p  p 1 ,…, p N . The system Hamiltonian can be expressed as:

H ðr,pÞ = K ðpÞ + V ðr Þ, ð20Þ

where K(p) and V(r) are the system kinetic and total potential energies, respectively. Then, the potential energy could
be decomposed into the following terms:

V ðr Þ = V P,b ðr P Þ + V L,b ðr L Þ + V E,b ðr E Þ + V PP,nb ðr P Þ + V LL,nb ðr L Þ + V EE,nb ðr E Þ + V PL,nb ðr PL Þ + V PE,nb ðr PE Þ + V LE,nb ðr LE Þ

ð21Þ

where VP,b, VL,b, and VE,b are the bonded potential energies in protein P, ligand L, and environment E, respectively.
VPP,nb, VLL,nb, and VEE,nb are the self nonbonded potential energies in protein P, ligand L, and environment E,
respectively. VPL,nb, VPE,nb, and VLE,nb are the nonbonded interaction energies between P–L, P–E, and L–E, respectively.
According to molecular mechanics force fields,105,106 the nonbonded potential energies are usually calculated as:

V nb = V elec + V vdW , ð22Þ

where Velec and VvdW are the system electrostatic and van der Waals potential energies. Presumably, ligand binding
mainly involves the nonbonded interaction energies of the ligand, VL,nb(r) = VLL,nb(rL) + VPL,nb(rPL) + VLE,nb(rLE). There-
fore, we add a boost potential selectively to the ligand non-bonded potential energy according to the GaMD algorithm:
8 9
< 1 k ðE 2 =
L,nb L,nb − V L,nb ðr ÞÞ , V L,nb ðr Þ < E L,nb
ΔV L,nb ðr Þ = 2 ð23Þ
: ;
0, V L,nb ðr Þ ≥ E L,nb

where EL,nb is the threshold energy for applying boost potential and kL,nb is the harmonic constant. These parameters
in LiGaMD are derived similarly as in the GaMD algorithm.
Next, one can add multiple ligand molecules in the solvent to facilitate ligand binding to proteins in MD simula-
tions. This is based on the fact that the average ligand unbound time τU is inversely proportional to the ligand concen-
tration [L], that is, τU = kon1½L with kon being the ligand binding rate constant. The higher the ligand concentration, the
faster the ligand binds, provided that the ligand concentration is still within its solubility limit. In addition to selectively
8 of 32 WANG ET AL.

boosting the bound ligand, another boost potential could thus be applied on the unbound ligand molecules, protein,
and solvent to facilitate both ligand dissociation and rebinding. The second boost potential is calculated using the total
system potential energy other than the non-bonded potential energy of the bound ligand as:
8
<1
k ðE −V D ðr ÞÞ2 , V D ðr Þ < E D
ΔV D ðr Þ = 2 D D ð24Þ
:
0, V D ðr Þ ≥ ED

where VD is the total system potential energy other than the nonbonded potential energy of the bound ligand, ED is the
corresponding threshold energy for applying the second boost potential and kD is the harmonic constant. This leads to
dual-boost LiGaMD (LiGaMD_Dual) with the total boost potential ΔV(r) = ΔVL,nb(r) + ΔVD(r).

2.5 | Peptide Gaussian accelerated molecular dynamics

Large conformational changes of peptides often occur via peptides binding to the target proteins, being distinct from
small-molecule ligand binding or protein–protein interactions (PPIs). We have developed another algorithm called pep-
tide GaMD or “Pep-GaMD” that enhances sampling of peptide–protein interactions.62
In Pep-GaMD, we consider a system of ligand peptide L binding to a target protein P in a biological environment E.
We decompose the potential energy into similar terms as in Equation (21). Presumably, peptide binding mainly involves
both the bonded and nonbonded interaction energies of the peptide since peptides often undergo large conformational
changes during binding to the target proteins. Thus, the essential peptide potential energy is VL(r) = VLL,b(rL) + VLL,
nb(rL) + VPL,nb(rPL) + VLE,nb(rLE). In Pep-GaMD, we add boost potential selectively to the essential peptide potential
energy according to the GaMD algorithm:
8
<1
k ðE −V L ðr ÞÞ2 , V L ðr Þ < E L
ΔV L ðr Þ = 2 L L ð25Þ
:
0, V L ðr Þ ≥ E L

where EL is the threshold energy for applying boost potential and kL is the harmonic constant.
In addition to selectively boosting the peptide, another boost potential is applied on the protein and solvent to
enhance conformational sampling of the protein and facilitate peptide rebinding. The second boost potential is cal-
culated using the total system potential energy other than the peptide potential energy as:
8
<1
k ðE −V D ðr ÞÞ2 , V D ðr Þ < E D
ΔV D ðr Þ = 2 D D ð26Þ
:
0, V D ðr Þ ≥ E D

where VD is the total system potential energy other than the peptide potential energy, ED is the corresponding threshold
energy for applying the second boost potential and kD is the harmonic constant. This leads to dual-boost Pep-GaMD
(Pep-GaMD_Dual) with the total boost potential ΔV(r) = ΔVL(r) + ΔVD(r).

3 | A P PL I C A T I O N S

Without the need to set predefined reaction coordinates or CVs, GaMD enables a wide range of applications
in enhanced sampling of biomolecules. Furthermore, accurate reweighting using cumulant expansion to the
second order could be achieved in GaMD simulations because the boost potential exhibits a Gaussian distribu-
tion, allowing recovery of the original free energy landscapes even for large biomolecules.47,50,51 Depending
on the system size, orders of magnitude speedup for biomolecular simulations could be achieved in GaMD. As
demonstrated on alanine dipeptide, GaMD simulations achieved 36–67 times speedup for sampling of the
backbone dihedral transitions compared with the long cMD simulations. 107 Higher acceleration could be
potentially achieved for larger systems with greater boost potential applied in the GaMD simulations.
WANG ET AL. 9 of 32

Hundreds-of-nanosecond to microsecond GaMD simulations could capture millisecond timescale events. Here,
we summarize recent application studies of GaMD.

3.1 | Protein–nucleic acid interactions

CRISPR–Cas9 system is a bacterial immune system that has introduced a powerful genome-editing technology, which
has revolutionized life sciences.108 At the molecular level, CRISPR-Cas9 is a protein/nucleic acid complex, composed of
the Cas9 protein associated with a guide RNA and matching sequences of DNA.109 Cas9 site-specifically recognizes the
DNA by binding its protospacer-adjacent motif (PAM), a short trinucleotide that enables the selection of the DNA
across the genome. Upon PAM binding, the DNA binds Cas9 by matching the RNA with one strand (the target strand,
TS), such forming an RNA:DNA heteroduplex structure. The second nontarget strand (NTS) of the DNA gets displaced
and also accommodated within the protein. Structures of the Streptococcus pyogenes Cas9 (SpCas9) revealed a bilobed
architecture (Figure 2(a)). One lobe—namely, the recognition lobe (REC)—includes three regions that mediate nucleic
acid binding (REC1–3), while the second is the nuclease lobe (NUC).110–112 The latter comprises two catalytic domains,
HNH and RuvC, which cleave the DNA TS and NTS, respectively. X-ray crystallography and cryo-EM studies portrayed
the structure of SpCas9 in different states, as apo protein,112 in complex with RNA113 and upon DNA binding.110,111
These structural studies have been a stepping stone to understand the mechanism of action of CRISPR–Cas9. However,
although critical, this information could not access the dynamics and the complex conformational transitions of this
ribonucleoprotein, raising fundamental questions on the system's biological function. In this regard, we have

F I G U R E 2 (a) Overview of the Streptococcus pyogenes CRISPR-Cas9 system. The Cas9 protein is represented in molecular surface,
showing individual domains in different colors. The RNA (yellow), target DNA (TS, violet), and nontarget DNA (NTS, cyan) are also shown.
(b) Energetic landscape associated with the conformational transition of the Cas9 protein from the apo form to the RNA-bound state,
computed using GaMD. The potential of mean force (PMF), which describes the free energy landscape, was computed along the E945–D435
FRET distance and the root mean square deviation (RMSD) with respect to the apo state. The simulations identified three minima: M1
corresponding to the crystallographic apo, M2 is the RNA-bound structure, and M3, which is an intermediate characterized by the solvent
exposure of an arginine-rich helix. For selected states, the ensemble-averaged electrostatic potential has been computed, revealing the
formation of a positively charged cavity (blue) in the intermediate states. Adapted with permission from Palermo et al. (2017). https://www.
pnas.org/content/114/28/7260. Copyright 2017 National Academy of Sciences
10 of 32 WANG ET AL.

successfully applied GaMD to decipher the molecular mechanism of nucleic acid processing and selectivity of this
genome-editing tool.

3.1.1 | Conformational changes underlying RNA binding to CRISPR–Cas9

Based on structural data, large structural transitions of the protein have been hypothesized to enable RNA bind-
ing.112,113 To characterize this process, we applied a GaMD in combination with Targeted MD (TMD) approach, which
reduces the RMSD between an initial and final target conformations.114 Using TMD, we obtained an initial pathway of
the conformational change from the apo protein to the RNA-bound form. We observed that the REC1–3 regions of the
protein moved in opposite directions relative to each other, leading to the closure of the REC lobe to accommodate
RNA. This observation agreed well with previous hypotheses based on cryo-EM.112,113 Then, we used GaMD to pre-
cisely describe the energetic landscape associated with this conformational change (Figure 2(b)). The free energy land-
scape described three local minima: M1 corresponds to the crystallographic apo structure, M2 is the RNA-bound
structure, while M3 is an intermediate state, characterized by the solvent exposure of an arginine-rich helix. The latter
directly binds the RNA guide in both RNA-bound and DNA-bound structures of Cas9,113 suggesting a mechanism for
the recruitment of RNA, in which the electrostatics could play a key role. By further computing the ensemble-averaged
electrostatic potential, we found that a positively charged cavity is formed at the level of the arginine-rich helix and is
suitable for RNA binding (Figure 2(b)). Overall, these simulations indicated that the arginine helix is critical for the
recruitment of RNA, and that the formation of positively charged cavity allows for the formation of the Cas9:RNA
binary complex.

3.1.2 | Conformational activation of the Cas9 protein for DNA cleavage

The process of conformational activation of the Cas9 protein toward DNA cleavages involves a critical transition of the
catalytic HNH domain. The latter undergoes a structural change from an inactive form to the active state prone to per-
form DNA cleavages.110,111,115 Our first study based on cMD highlighted a “striking plasticity” of the catalytic HNH
nuclease.116 That study revealed the critical dynamic interplay between the DNA NTS and HNH, suggesting that the
binding of the NTS would allow increased dynamics of HNH and, thereby, its activation toward DNA cleavage. After
this first computational study, single-molecule Förster Resonance Energy Transfer (FRET) experiments have investi-
gated the large scale dynamics of the system, showing that the dynamical docking of HNH at the cleavage site critically
requires the presence of the NTS,117 and thereby confirming the predictions of molecular simulations. Overall, that
early MD study has been instrumental in characterizing the atomic-level details of the CRISPR–Cas9 dynamics. Never-
theless, considering that cMD simulations are limited to short timescales (i.e., ns–to–μs), that investigation could not
fully address the activation mechanism of the catalytic HNH domain. To overcome the inherent timescale limits of
cMD and characterize the HNH activation process, we performed GaMD simulations, capturing multiple states of the
HNH conformational landscape.114 As expected, the simulations broadly sampled various possible configurations of the
HNH domain. Energetic reweighting of the conformational landscape revealed that the energetic minima identified
through GaMD correspond to the conformational states found by FRET experiments111,113,115,118 and structural stud-
ies110 (Figure 3(a)). Notably, this extensive GaMD sampling (collecting >20 μs) identified a “bona-fide” conformation of
the active state (namely, active #4 in Figure 3(a)), which was shown to be thermodynamically stable. This conformation
predicted the active state 2 years before structural data was made available,110 allowing also to start in-depth studies of
the catalysis through hybrid quantum-classical methods.119,120 Moreover, to better describe the conformational change
from the pre-active state (captured in PDB ID: 5f9r) to the active configuration where HNH catalytic residue (H983)
docks at the DNA TS (Figure 3(b), top panel), the specialized supercomputer Anton-2121 has been used to carry out con-
tinuous multi-μs cMD simulations.122 These simulations captured the late step of HNH activation over 16 μs of con-
tinuous simulations (Figure 3(b), bottom panel). The dynamical docking of HNH at the cleavage site on the TS
occurred by following the same pathway previously observed over multiple GaMD replicas (Figure 3(b), central panel).
Indeed, while the continuous simulation performed on Anton-2 recovered the transition over 16 μs, GaMD captured
the conformational change by running 400 ns and in three replicas. This finding indicates that GaMD reliably cap-
tures structural transitions of biomolecules that occur over longer time scales. Finally, it is notable that the activated
state, which was early identified through GaMD114 and later refined using Anton-2 simulations,122 resulted in notable
WANG ET AL. 11 of 32

agreement with the cryo-EM structure of the active complex.110 This showed the reliability of the early predictions that
have been obtained based on GaMD.

3.1.3 | Molecular mechanism of off-target effects of CRISPR–Cas9

An important mechanistic question relates to the onset of off-target effects, which arise from the binding of DNA
sequences that do not fully match the guide RNA, resulting in RNA:DNA hybrids containing mismatched pairs. Off-
target effects result in cleavages at DNA sites, representing a limitation for the application of CRISPR-Cas9 for in vivo
and ex vivo genome editing. Kinetic and single-molecule FRET studies provided critical hints on the molecular basis of
off-target effects. Indeed, it has been shown that DNAs containing 1–3 mismatches located at the RNA:DNA hybrid
ends result in a flexible and catalytically active HNH domain.117,123,124 Contrariwise, four (or more) mismatches result
in decreased flexibility of HNH and in its catalytic inactivation. The single-molecule experiments, however, could not
explain how a different number of DNA mismatches at the RNA:DNA hybrid ends could affect the activation of HNH.

F I G U R E 3 (a) Conformations of the HNH domain (green) in its inactive (#1, #2), pre-active (#3), and active (#4) states, as
experimentally determined through single-molecule FRET and structural approaches (top panel). The free energy landscape (i.e., potential of
mean force, PMF) associated with the conformational changes of the HNH domain from its inactive to active states is shown in the bottom
panel. The minima correspond to the four states experimentally found (top). The PMF was computed along with the S867-S355 and
N1054-S867 FRET distances. Adapted with permission from Palermo et al. (2017). https://www.pnas.org/content/114/28/7260. Copyright
2017 National Academy of Sciences. (b) Conformational change of the HNH domain from its pre-active conformation (captured in the PDB
ID: 5F9R, left) to the active state identified through GaMD (right). The active state displays the catalytic residue H983 close to the DNA
target strand (TS). The distance between the catalytic H840 and the scissile phosphate (H840–PDNA) has been computed along 400 ns
GaMD (central panel) and 16 μs of continuous MD using the specialized supercomputer Anton-2 (bottom panel). The black dashed line
indicates the pre-active conformation (PDB ID: 5F9R) used as a starting point for MD simulations, while the magenta dashed line indicates
the active conformation more recently captured through cryo-EM (PDB ID: 6O0Y). Reprinted with permission from Palermo et al. (2018).
Copyright 2018 Cambridge university press. https://doi.org/10.1017/S0033583518000070
12 of 32 WANG ET AL.

Knowing the molecular basis of this mechanism is of critical importance, as it could help in developing more specific
CRISPR–Cas9 systems, in which a single basepair mismatch is sufficient for reducing the HNH dynamics and catalytic
function, thereby inhibiting the cleavage of incorrect DNA sequences. Considering that GaMD has been successful in
describing the activation mechanism of HNH114 (Figure 3), we used the method to investigate the effect of base pair
mismatches on its conformational dynamics.67,125 For this application, we employed GaMD without carrying out ener-
getic reweighting, increasing the sampling of low-energy states and providing a semi-quantitative ranking of the associ-
ated probabilities. This enabled us to broadly explore the system's conformational dynamics in the presence of base pair
mismatches. The simulations revealed that four or more mismatches induce a broad opening of the RNA:DNA hybrid
(Figure 4(a)), which results in newly formed interactions between the TS and the L2 loop. These interactions impor-
tantly reduce the HNH flexibility, hampering its conformational activation. On the other hand, 1–3 base pair mis-
matches do not result in sensible openings of the heteroduplex, as evinced by the minor groove width (measured at
position 17, Figure 4(b)), resulting in a negligible effect on the HNH conformational dynamics and thereby not affecting
its activation for cleavage. Overall, the simulations could discriminate the different effects of base pair mismatches on
the HNH activation, providing a mechanistic rationale to previous kinetic and single-molecule experiments. Building
on the outcomes of GaMD simulations, we suggested that altering the TS–L2 interactions could reduce off-target bind-
ing. This speculation has been supported by the experimental engineering of the L2 loop in several variants of the Cas9
enzyme,124,126,127 which increase the system's specificity toward on-target sequences.

3.1.4 | Allosteric effects across the CRISPR–Cas9 complex

Multiple evidences including experiments and computations have indicated that CRISPR–Cas9 is also an intriguing
“allosteric engine”.124,128–130 Indeed, CRISPR–Cas9 requires an intricate allosteric activation to accomplish DNA cleav-
ages. Biochemical experiments have indicated that the central element of the CRISPR-Cas9 allosteric signaling is the
HNH domain, since its high flexibility can allow the signal transmission. To describe the allosteric signaling across
HNH and how it transfers the information of DNA binding (occurring within the REC lobe) to the catalytic sites for
cleavage, GaMD was combined with graph theory.131 This combination allowed inclusion of long-timescale motions in
the calculation of the allosteric pathways.66 Specifically, while GaMD characterized the long timescale system's dynam-
ics, network models derived from graph theory accurately described the allosteric network and information transfer.
This approach revealed the existence of a millisecond timescale dynamic pathway across HNH, which connects the
RuvC nuclease domain to the recognition lobe REC. This allosteric route was validated through NMR relaxation experi-
ments, showing that a contiguous pathway of slow residues overlaps with the prediction from GaMD and graph theory-
based analysis. In summary, the combination of GaMD simulations with graph theory provided a useful approach for

F I G U R E 4 (a) Extended opening of the RNA:DNA hybrid and newly formed interactions with the L2 loop (magenta) of the HNH
domain (green), observed during GaMD simulations of CRISPR-Cas9 in the presence of four base pair mismatches at the RNA:DNA hybrid
ends. (b) RNA:DNA minor groove width computed along with MD simulations of CRISPR-Cas9 bound to an on-target DNA (black) and in
the presence of 1–4 mismatches at the hybrid ends. A vertical bar indicates the experimental minor groove width (i.e., 11 Å from X-ray
crystallography). The minor groove width has been measured at the level of basepair 17 (shown on the right). Adapted with permission from
Ricci et al. (2019). Copyright 2019 American Chemical Society. https://pubs.acs.org/doi/full/10.1021/acscentsci.9b00020. Further permissions
related to the material excerpted should be directed to the American Chemical Society
WANG ET AL. 13 of 32

determining the signal transduction in CRISPR-Cas9, laying the foundations for characterizing allostery in other pro-
tein/nucleic acid complexes whose biological function relies on slow dynamical motions associated with the (re)organi-
zation of protein domains and long-range effects.

3.2 | Protein–protein/peptide interactions

Protein–protein interactions (PPIs) and protein-peptide interactions are central to biological functions and have thus
been targeted to design novel therapeutic drugs.132–136 Here, we will summarize recent GaMD applications in simula-
tion studies of PPIs and protein–peptide interactions.

3.2.1 | Protein–protein interactions

The recognition of T cell receptor (TCR) and peptides presented by major histocompatibility molecules (pMHC) ini-
tiates adaptive immune responses. The pMHC binding affinity often correlates with the TCR-signaling strength.
However, frequent high-affinity of pMHC in the human T-cell repertoire are not stimulatory. Recently, enhanced
sampling methods including GaMD, ABF, and SMD were performed to distinguish stimulators from nonstimulatory
ligands by simulating the TCR-pMHC disengagement.70 The GaMD was first performed to reveal the structural flexi-
bility of the complex and identify important CVs, including the orientation angle of the TCR about the pMHC
assembly, salt bridges, and hydrogen bonds. Then, the identified CVs were used for free energy calculation using
the ABF method. Constant velocity SMD simulations were performed starting from the free energy minima identi-
fied by ABF. The simulations revealed that dynamic interactions in the TCR–pMHC interface play a critical role in
determining the TCR specificity. One collective property of the entire TCR–pMHC interface is the formation of a
catch or slip bond, being consistent with the results from single-molecule force measurements. In addition to simu-
lations of PPIs with globular proteins such as the TCR-pMHC complex, GaMD has also been successfully applied to
investigate PPIs with membrane proteins such as GPCR-G protein interactions,68,69 which will be described in Sec-
tion 3.5. In summary, GaMD is suitable to study large biomolecular complexes and provides important insights into
functionally important PPIs.

3.2.2 | Protein–peptide interactions

Petrizzelli et al.72 applied GaMD to investigate the pathogenic mechanisms caused by missense mutations of KDM6A
on the histone H3, including P941S, D980V, S1025G, H1060L, L1200F, G1223D, Q1248R, and R1255W. GaMD simula-
tions showed that the interaction between the linker and JmjC domains was significantly impacted by residue muta-
tions, leading to a loss of function. All mutants exhibited movements of the disordered linker domain, leading to the
increased flexibility of the KDM6A–H3 complex, which induced wrong exposure and orientation of the trimethylated
lysine in the catalytic site. Therefore, GaMD simulations revealed important pathogenic mechanisms of the KDM6A–
H3 interaction.72
We developed a novel approach, namely, PeptiDock + GaMD, in which the global peptide docking-ClusPro Pep-
tiDock and GaMD simulations were combined for improving modeling of protein–peptide interactions.137 For three
model peptides (peptide 1–3), docking models generated with PeptiDock138 showed 3.3, 3.5, and 4.8 Å RMSD of the pep-
tide backbone relative to their experimental structures. The peptide docking poses were refined by GaMD simulations.
Then, the PyReweighting toolkit42 was applied to reweight and calculate free energies of the peptide structural clusters
obtained from GaMD simulations. RMSDs of Peptides 1 and 2 in the 1st top-ranked cluster were 0.9 and 0.6 Å, respec-
tively. The third top-ranked cluster in Peptide 3 exhibited the smallest RMSD of 2.7 Å. Thus, the PeptiDock + GaMD
could be used to accurately predict the peptide–protein interaction. In comparison, cMD simulations with the same
simulation time were much less efficient in refining the peptide docking poses.138 Only one among four cMD simula-
tions of Peptide 2 improved the peptide binding pose. RMSD decrease was not observed in any of cMD simulations of
Peptide 3. The top-ranked models obtained by clustering of cMD snapshots were of high quality for only Peptide 1 but
medium for both Peptides 2 and 3. Therefore, GaMD simulations refined peptide docking poses and provided signifi-
cantly improved sampling than cMD.
14 of 32 WANG ET AL.

The PeptiDock + GaMD approach was further applied to model interactions of cyclic peptides with proteins,
including peptide binding to MDM2/MDMX139 and Tsg101 UEV protein.71 Compared with the linear peptides, cyclic
peptides often possess longer lifetime and better biological activity. To facilitate cyclic peptide design, the PeptiDock
+ GaMD approach was applied to investigate binding interactions between the UEV domain protein and three cyclic
peptides.71 The predicted peptide binding mode identified from GaMD simulations was further validated by binding
free energy calculations, which agreed well with the experimental binding affinities. Therefore, GaMD simulations
provided important insights into protein–peptide interactions and were applicable to both linear and cyclic peptides.

3.2.3 | Binding thermodynamics and kinetics of peptide

Pep-GaMD62 has been developed to simulate both peptide binding and dissociation, which allows us to calculate the
binding free energies and kinetics of flexible peptides. It has been demonstrated on binding of three model peptides to
the SH3 domains,140,141 which include “PPPVPPRR” (PDB: 1CKB), “PPPALPPKK” (PDB: 1CKA), and “PAMPAR”
(PDB: 1SSH) (Figure 5(a–c)). Repetitive peptide binding and unbinding events were captured in independent 1 μs Pep-
GaMD simulations, allowing us to calculate peptide-binding thermodynamics and kinetics (Figure 5(d–i)). Peptide
kinetics especially the dissociation rate was accelerated by 3–4 orders of magnitude in the Pep-GaMD simulations.
The predicted values from Pep-GaMD were in good agreement with available experimental data. Furthermore, the Pep-
GaMD simulations revealed the important role of long-range electrostatics in peptide binding and the binding mainly
followed a conformational selection model.

3.3 | Protein–ligand binding

3.3.1 | Protein–ligand interactions

Use of EGFR tyrosine kinase inhibitors is a promising approach to improve progression-free survival in cancer
patients. Osimertinib was approved as a third-generation EGFR mutant selective inhibitor. Unfortunately,
resistances were detected against the osimertinib therapy. Brown et al. 79 identified G724S as an osimertinib
resistance mutation. Microsecond GaMD simulations were performed on EGFR mutants in the presence and
absence of osimertinib to investigate the underlying mechanism. The GaMD simulations showed that the
G724S mutation disrupts the osimertinib binding to the enzyme with exon 19 in-frame deletion (Ex19Del)
mutation, while does not affect the enzyme with exon 21 missense mutation (L858R). The G724S mutation
induces hyper stabilization of glycine-rich P-loop in β-bend conformation. It disrupts the interaction between
the indole ring of osimertinib and phenyl ring of F723. These results were further verified in animal cell cul-
ture experiments and in cancer patients. Overall, GaMD simulations elucidated the molecular mechanisms of
ligand binding in EGFR mutations for treatment of nonsmall cell lung carcinomas, as well as many other
protein–ligand interactions.55,56,73–76

3.3.2 | Ligand binding thermodynamics and kinetics characterized by LiGaMD

LiGaMD has been proposed to quantitatively characterize ligand binding thermodynamics and kinetics.61 Host–guest
and protein–ligand binding model systems have been used to validate the LiGaMD algorithm. Hundreds-of-nanosecond
LiGaMD simulations captured repetitive guest binding and unbinding in the β-cyclodextrin host. The calculated guest
binding free energies were in good agreement with experimental data, for which the errors were <1.0 kcal/mol. The
sampling errors of LiGaMD simulations were <1.0 kcal/mol in comparison with converged μs-timescale cMD simula-
tions. Additionally, ligand kinetic rate constants were accurately predicted using Kramers' rate theory. Furthermore,
repetitive binding and unbinding of the benzamidine inhibitor in trypsin was observed in 1 μs LiGaMD simulations,
allowing us to accurately calculate ligand binding free energy and kinetic rate constants. The ligand dissociation rate
was remarkably accelerated by seven orders of magnitude in the LiGaMD simulations. The predicted values were in
excellent agreement with the experimental data.61
WANG ET AL. 15 of 32

F I G U R E 5 Pep-GaMD simulations have captured repetitive dissociation and binding of three model peptides to the SH3 domains: (a–c)
X-ray structures of the SH3 domains bound by peptides (a) “PAMPAR” (PDB: 1SSH), (b) “PPPALPPKK” (PDB: 1CKA), and (c) “PPPVPPRR”
(PDB: 1CKB). The SH3 domains and peptides are shown in green and magenta cartoons, respectively. Key protein residues Asp19 and Trp40
in the 1SSH structure and Asp150 and Trp169 in the 1CKA and 1CKB structures, and peptide residues Arg10 in the 1SSH structure, Lys8 in
the 1CKA structure, and Arg7 in the 1CKB structure are highlighted in sticks. The “N” and “C” labels denote the N-terminus and C-
terminus of the peptides. (d–f) Time courses of peptide backbone RMSDs relative to X-ray structures with the protein aligned calculated
from three independent 1 μs Pep-GaMD simulations of the (d) 1SSH, (e) 1CKA, and (f) 1CKB structures. (g–i) The corresponding PMF
profiles of the peptide backbone RMSDs averaged over three Pep-GaMD simulations of the (g) 1SSH, (h) 1CKA, and (i) 1CKB structures.
Error bars are standard deviations of the free energy values calculated from three Pep-GaMD simulations. Reprinted from Wang and Miao
(2020), with the permission of AIP Publishing

3.4 | Protein enzymes

3.4.1 | Structural dynamics of protein kinases

Casein kinase 1δ (CK1δ) has been regarded as an important component in metazoan circadian rhythms regulation.
Despite its importance, little was known about substrate selectivity and activity of the enzyme in molecular detail.
16 of 32 WANG ET AL.

Philpott et al.89 performed GaMD simulations on wildtype and tau mutant CK1δ systems and discovered a conforma-
tional switching mechanism of the activation loop. The switch regulates two different regions of the PER2 protein,
which in turn regulates the protein stability and circadian timings in eukaryotes. The GaMD simulations further rev-
ealed that anion binding to a highly conserved site monitors the conformation in the activation loop and thereby regu-
lating the overall conformation of the substrate-binding cleft. The tau mutant, on the other hand, disrupts the allosteric
regulation between the anionic sites. This disturbs the conformational flexibility of the activation loop and affects the
stability of the PER2 protein. GaMD simulations thus provided molecular basis of the decreased activity in the tau
mutant CK1δ.
Brassinosteroid insensitive 1-associated kinase 1 (BAK1) is an important receptor-like kinase which initiates
numerous immune and growth signaling pathways in plants. Moffett et al.88 applied GaMD simulations to explore
physiochemical basis of BAK1 activation through phosphorylation. GaMD simulations revealed the effects of various
phosphorylation patterns and ATP binding on the enzyme conformation. GaMD simulations identified a metastable
inactive enzyme conformation using activation-loop cracking. This activation loop conformation had been also found
in other kinases like the ERK2. Phosphorylation of residues T450 and T455 played important roles in stabilizing the
active-like activation loop without cracking. During the GaMD simulations, phosphorylation helped αC helix of the
enzyme maintain its position near the N-lobe. In contrast, the αC helix of the unphosphorylated systems switched to
an inactive state as the activation loop changed into a “cracked” conformation. Overall, GaMD simulations revealed
the mechanism of phosphorylation controlled BAK1 activation. In another study, Koh et al.87 performed GaMD sim-
ulations for mechanistic insights into the flux-dependent transport signaling by Bce-like antibiotic resistance systems.
They found that the transport activity is directly related to histidine kinase activity even with different antibiotic
concentrations.

3.4.2 | Active site dynamics of protein enzymes

The oncoprotein AlkB homolog 5 (Alkbh5) is involved in cancers such as leukemia, brain cancer, and breast cancer.
NMR experiments and GaMD simulations were combined to generate the structural model of the apo human Alkbh5.80
The Alkbh5 active site was observed to be more disordered than that in the x-ray structure (PDB:4NJ4). It was likely
due to the absence of the Cys230–Cys267 disulfide bond in solution, which limited the protein conformational accessibil-
ity. GaMD simulations captured breathing motions of the protein, which expands the α-ketoglutarate binding pocket
and permits binding of small molecules.
FabA and FabZ are two Escherichia coli dehydratases involved in the production of the unsaturated fatty acids
(UFAs) from fatty acid biosynthesis.82 Both FabA and FabZ are known to catalyze dehydration reactions, however,
only FabA can further catalyze isomerization reaction. A combined approach involving chemical biology, structural
biology, and GaMD simulations was applied to understand the substrate selectivity and divergent activity of the two
enzymes. Cross-linking experiments were performed to produce the acyl-AcpP•FabA and acyl-AcpP•FabZ com-
plexes, which were used for GaMD simulations to elucidate the dehydration mechanism catalyzed by FabA and
FabZ. GaMD simulations revealed the dynamic mechanism of the unique isomerase activity of FabA and successfully
differentiated the substrate preferences of FabA and FabZ. Moreover, GaMD simulations showed that only FabA
selectively sampled the (−) gauche conformer of trans-2-decenoyl-AcpP for allylic rearrangement.
Furthermore, GaMD was successfully applied on simulations of protein enzymes for structure-based drug design of
anti-malarial drugs,81 design of inhibitors targeting Staphylococcus aureus enzyme MnaA,84 investigations of soybean
lecithin–gallic acid complex formation to aid in alcoholic liver disease (ALD),86 usefulness of antioxidative agent for
treating vascular endothelial deficits,85 and understanding of drug resistance mechanism of rifampin.91 Simulations
using GaMD and replica-exchange solute tempering (REST2)142 were performed to understand the mechanism of the
transactivation of estrogen receptor.78

3.4.3 | Protein allostery

The interaction between HCV NS5A-D2 and human prolyl isomerase cyclophilin A (CypA) plays an essential role in
viral RNA replication. Dujardin et al.83 employed GaMD simulations and NMR to investigate the role of a short struc-
tural motif PW-turn (314PXWA317) on the structural disorder in NS5A-D2. There is a conformational equilibrium
WANG ET AL. 17 of 32

between folded and disordered states in the PW-turn motif, which is allosterically regulated by the cis/trans isomeriza-
tion of 5 proline residues (P306, P310, P315, P319, and P320). Moreover, the HCV RNA replication efficiency correlates
well with the fraction of the structured PW-turn obtained from GaMD simulations.
Another study by Sztain et al.90 employed GaMD to identify cryptic pockets of the SARS-CoV-2 main protease
(Mpro), which are far away from the active site. Four systems including the monomer and dimer of Mpro in the
absence and presence of the co-crystalized N3 inhibitor were built to perform GaMD simulations. Three regions includ-
ing the distal allosteric site, active site, and dimer interface region were identified as potential drug pockets using the
PockDrug webserver.143 Virtual screening against the above-mentioned pockets allowed to identify more hit molecules
than using only the active site in the crystal structure. Furthermore, correlation analysis suggested that the three
pockets could be allosterically regulated by each other. Therefore, the above-identified pockets could be useful in virtual
screening to identify novel inhibitors of SARS-CoV-2.

3.5 | Membrane proteins

Membrane proteins including GPCRs, intramembrane proteases, and ion channels play essential roles in cellular sig-
naling and serve as important drug targets. Here, we will summarize recent applications of GaMD in studies of GPCRs
(including muscarinic acetylcholine, adenosine, opioid, and chemokine receptors), γ-secretase, and so forth.

3.5.1 | Binding mechanism of G protein mimetic nanobody to M2 muscarinic GPCR

GaMD simulations were performed to capture the spontaneous binding of nanobody Nb9-8, a G protein mimetic, to the
M2 muscarinic GPCR.144 The agonist and nanobody in the X-ray structure of the active M2 receptor were placed to be
>20 Å far away from the receptor to build the starting model. Five independent GaMD simulations lasting 4500 ns
were performed. One GaMD simulation successfully captured the binding of the nanobody to the receptor G-protein
coupling site of M2 with a minimum RMSD of 2.48 Å in the nanobody core domain relative to the X-ray conformation,
although the agonist still not reached its binding site (Figure 6(a,b)). Both the orthosteric ligand-binding pocket and
intracellular domains of the M2 receptor involved conformational change along with the binding of the nanobody
(Figure 6(b)). The orthosteric pocket in the X-ray structures of antagonist-bound and agonist nanobody-bound receptor
are “open” and “closed”, respectively. Binding of the nanobody induced the orthosteric pocket from the “open” to
“closed” state. Moreover, activation of the M2 receptor was occurred during the binding of the nanobody, as measured
by the distance between intracellular transmembrane helix 3 and 6 (TM3−TM6 distance; Figure 6(b)). Free energy pro-
file of the nanobody RMSD relative to the 4MQS X-ray conformation and the receptor Arg1213.50–Thr3866.34 distance
were calculated to characterize the nanobody binding pathways (Figure 6(c)). Three low-energy conformational states
including the unbound (U), intermediate 1 (I1), and intermediate 2 (I2) were identified from the potential of mean force
(PMF) profile. The bound (B) conformation identified in the GaMD simulations is similar to that sampled in previous
simulations of the 4MQS X-ray structure.144 On the intracellular side, the nanobody core domain especially the β2, β3,
β6, β7, and β8 strands overlapped well with the 4MQS X-ray structure when the nanobody RMSD decreased to 2.48 Å
(Figure 6(d)). Therefore, the GaMD provided important insights into the binding mechanism of the nanobody to the M2
receptor.

3.5.2 | Mechanism of specific G protein coupling to adenosine receptors

There are four subtypes of adenosine receptors (A1, A2A, A2B, and A3) in human, which mediate the effects of adenosine
(ADO). The odd ARs including A1AR and A3AR mainly couple to the Gi/o proteins, while the even ARs (A2AAR and
A2BAR) preferentially couple to the Gs proteins. In one of our recent studies,68 we used GaMD simulations on four
AR-G protein models, including the native structures of ADO-A1AR-Gi with ADO and Gi protein bound145 and the
NECA-A2AAR-Gs with 50 -N-ethylcarboxamidoadenosine (NECA) and an engineered Gs protein-bound,146 as well as
“decoy” complexes ADO-A1AR-Gs and NECA-A2AAR-Gi generated by switching the G proteins. In the ADO-A1AR-Gi
and NECA-A2AAR-Gs complexes, GaMD identified only one stable low-energy conformation, which is similar to the
cryo-EM structure (Figure 7(a,b)). Similarly, only one low-energy conformation was identified in the NECA–A2AAR–Gi
18 of 32 WANG ET AL.

F I G U R E 6 Binding of agonist IXO and Gi protein mimetic nanobody Nb9-8 to the M2 muscarinic GPCR was captured in one of five
GaMD simulations: (a) Trajectories of a nitrogen atom in IXO (beads) and the β8 strand of Nb9-8 (ribbons) colored by simulation time in a
blue (0 ns)–white (2250 ns)–red (4500 ns) scale. (b) RMSDs of the IXO and Nb9-8 relative to the X-ray structure, Tyr1043.33-Tyr4036.51-
Tyr4267.39 Triangle perimeter and Arg1213.50-Thr3866.34 Distance calculated from the simulation. Dashed lines indicate X-ray structural
values of the M2 receptor (3UON: Green and 4MQS: Red). (c) Binding pose of IXO (spheres) in the receptor extracellular vestibule with
13.84 Å RMSD relative to the X-ray conformation (yellow spheres). Residues found within 5 Å of IXO are highlighted in sticks. (d) Binding
of Nb9-8 (cyan), which exhibits only 2.48 Å RMSD in the protein core (the β2, β3, β6, β7, and β8 strands). X-ray conformations of the M2
receptor and nanobody are shown in orange and purple ribbons, respectively. Adapted with permission from Miao et al. (2018). https://
www.pnas.org/content/115/12/3036. Copyright 2018 National Academy of Sciences

complex, suggesting that agonist NECA binding in the A2AAR could be still stabilized by coupling to the Gi protein
(Figure 7(c)). While the binding of Gs protein in the A1AR led to increased fluctuations of both the receptor and ADO
(Figure 7(d)). The ADO agonist exhibited high fluctuations and sampled two different binding poses (L1 and L2) in the
WANG ET AL. 19 of 32

F I G U R E 7 2D potential of mean force (PMF) profiles of the (a) ADO-bound A1AR-Gi, (b) NECA bound A2AAR-Gs, (c) NECA bound
A2AAR-Gi, and (d) ADO bound A1AR-Gs complex systems regarding the agonist RMSD relative to the cryo-EM conformation and AR:
NPxxY-G:α5 distance. The white triangles indicate the cryo-EM or simulation starting structures. Summary of specific AR-G protein
interactions: (e) the ADO-bound A1AR prefers to bind the Gi protein to the Gs. The latter could not stabilize binding of agonist ADO in the
A1AR and tended to dissociate from the receptor. (f) The A2AAR could bind both the Gs and Gi proteins, which adopted distinct
conformations in the complexes. Adapted with permission from Wang et al. (2019). Copyright 2019 American Chemical Society. https://
pubs.acs.org/doi/10.1021/acs.jpcb.9b04867. Further permissions related to the material excerpted should be directed to the American
Chemical Society
20 of 32 WANG ET AL.

F I G U R E 8 GaMD simulations revealed the activation and its ε cleavage mechanisms of γ-secretase in the wildtype and mutant APP
substrates. Summary of the (a) inactive cryo-EM, (b) active (wildtype), and (c) shifted active (M51F) conformational states of the APP-bound
γ-secretase. Distinct AICD products were generated from the wildtype and M51F mutant APP. GaMD free energy profiles of (d) wildtype and
(e) M51F APP-bound γ-secretase regarding the Asp257:Cγ–Asp385:Cγ and Asp257:protonated O–Leu49:O distances. Adapted with
permission from Bhattarai et al. (2020). Copyright 2020 American Chemical Society. https://pubs.acs.org/doi/abs/10.1021/acscentsci.
0c00296. Further permissions related to the material excerpted should be directed to the American Chemical Society

ADO–A1AR–Gs complex. In the “L2” binding pose, ADO interacted with residues Tyr121.35 and Tyr2717.36 in the sub-
pocket 2 of the A1AR, which is described earlier.147 GaMD simulations indicated that coupling with the Gi protein was
preferred to the Gs in the A1AR (Figure 7(e)), while both the Gs and Gi proteins could be coupled with the A2AAR
(Figure 7(f)), being well agreement with experimental data of the ARs.148–150 In summary, the dynamic mechanism of
specific GPCR-G protein interactions could be obtained from the GaMD simulations.

3.5.3 | GPCR–membrane interactions depend on the receptor activation state

The phospholipid membrane bilayer plays an important role in GPCR transiting among different conformational states.
Elucidation of the lipid–protein interactions could facilitate to understand the functional mechanism of GPCRs. In one
of our recent study,95 the cryo-EM structure of the active ADO–A1AR–Gi145 and the X-ray structure147,151 of the inactive
antagonist PSB36-bound A1AR (PSB36-A1AR) were used to perform GaMD simulations. They were embedded in a
1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC) lipid bilayer. GaMD simulations revealed important role of the
membrane lipids in stabilizing different states of the A1AR. Different structural flexibility profiles of the inactive and
active A1AR were obtained by the GaMD simulations. In comparison with the inactive state, higher fluctuations of the
A1AR ECL2 region, intracellular ends of TM6 and TM5 were found in the active state. Furthermore, the -SCD order
WANG ET AL. 21 of 32

parameter values obtained from GaMD simulations were consistent with experimental data.152 Particularly, the inactive
and active A1AR systems exhibited similar values of the -SCD order parameters of sn-2 acyl chains of POPC in the upper
leaflet. However, the active A1AR exhibited smaller value of the same -SCD order parameters in the lower leaflet than
those in the inactive A1AR, suggesting that POPC lipids in the lower leaflet of the active A1AR system were more fluid
than in the inactive A1AR system. One reliable explanation is that the outward movement of TM6 in the active A1AR
could induce higher inclination of the C H bonds to be aligned along the bilayer normal. Thus, GaMD simulations
showed that the protein–membrane interactions depended on different conformational states of the A1AR.

3.5.4 | Mechanism of allosteric drug lead binding to an adenosine GPCR

Preclinical studies suggest that the A1AR is an important drug target for treating diseases including reduce neuropathic
pain and ischemia–reperfusion injury.153–155 However, off-target side effects have hindered the therapeutic develop-
ment of A1AR agonists, which mainly originated from the high conservation of the endogenous agonist binding
(orthosteric) site across different AR subtypes.156 Positive allosteric modulators (PAMs), which bind to a less conserved
“allosteric” site, have the potential to develop high subtype-selective A1AR therapeutics.157 Using the X-ray structure of
the A1AR (PDB: 5UEN)147 as a model, GaMD simulations158 were performed to investigate binding mechanisms of two
PAMs, VCP171, and PD81723. Each PAM was initially placed at >20 Å away from the receptor. Spontaneous binding
of PAMs to the A1AR was captured in GaMD simulations using both AMBER47 and NAMD51 at different acceleration
levels (dihedral and dual boost).
GaMD simulations identified similar binding modes of PD81723 and VCP171 that bound to a site around ECL2 in
the A1AR from different acceleration levels of GaMD simulations performed using different software packages. They
were in highly agreement with experimental results of site-directed mutagenesis obtained by the Christopoulos
group.159 In the mutagenesis experiments, numerous ECL2 residues including Asn148ECL2, Glu153ECL2, Ser161ECL2,
Ile167ECL2, and Glu172ECL2 were mutated to alanine, which caused significant alterations in PAM binding affinity, effi-
cacy, and cooperativity.159,160 These residues were found to interact with the bound PAMs in the GaMD simulations.
Additionally, agonist binding affinity was enhanced by PAM binding. In the absence of PAM, the agonist sampled a
large conformational space in the receptor orthosteric pocket without binding of PAM. Upon PAM binding at the
ECL2, movement of agonist was significantly reduced.

3.5.5 | Mechanism of drug binding to a chemokine GPCR

Chemokine receptors are regarded as one of the important GPCRs with implications in human health and therapeutics.
CXCR4 is an important subtype with involvement in different human diseases including cancer and HIV infection.
Despite its importance, less is known about the mechanism of drug interaction with the receptor. GaMD simulations73
were performed to study the binding mechanism of the drug Plerixafor (PLX) and its pathway to CXCR4. Simulation
systems were built by placing 10 unbound ligand molecules at a distance >15 Å away from the receptor.
GaMD_Dual_NB boost scheme was used in which system nonbonded and dihedral energy terms were boosted. The
GaMD_Dual_NB simulations captured spontaneous binding of the PLX from the bulk solvent to the receptor
orthosteric site in one of five production runs. The complete binding of PLX was observed at 480 ns timescale with
minimum RMSD relative to the bound conformation of 2.76 Å. In the binding pocket of CXCR4, the positively charged
PLX formed stable salt bridges with residues Asp972.63, Asp2626.58, and Glu2887.39 occupying both the minor and major

TABLE 1 The implemented GaMD algorithms in different MD software packages

AMBER CPU version AMBER GPU version NAMD Genesis

Dihedral GaMD X X X X
Total boost GaMD X X X X
Dual boost GaMD X X X X
Nonbonded dual-boost GaMD X X
Selective GaMD (LiGaMD and Pep-GaMD) X
22 of 32 WANG ET AL.

sub-pockets of the receptor. In 2D PMF profile calculated from the GaMD simulations, “unbound”, “intermediate 1”
(I1), “intermediate 2” (I2), and “bound” PLX low-energy conformational states were identified. In the intermediate con-
formational states I1 and I2, same polar and charged residues in the receptor ECL2–TM5–TM6 region, namely
Asp187ECL2, Asp1935.32, and Asp2626.58, formed favorable interactions with the positively charged nitrogen atoms of
PLX. Thus, the ECL2–TM5–TM6 region of CXCR4 formed a novel intermediate drug binding site. Furthermore, GaMD
simulations identified PLX drug binding pathway to the CXCR4. These studies are expected to greatly facilitate the drug
design of CXCR4.

3.5.6 | Mechanisms of γ-secretase activation and substrate processing

The mechanism of activation of γ-secretase bound to amyloid precursor protein (APP) was investigated using GaMD
simulations.97 The Cryo-EM structures of two substrates App and Notch bounded γ-secretase were simulated to investi-
gate substrate processing by γ-secretase of wildtype and APP mutant causing the familial Alzheimer's disease (FAD)161
(Figure 8(a)). Mutations in the cryo-EM structure introduced unnatural enzyme-substrate interactions and hindered
the activation process. With a combined study of GaMD simulations, mass spectrometry, and western blotting, a model
highlighting the process of intramembrane proteolysis of APP by γ-secretase was presented.
Spontaneous activation of γ-secretase in complex with wildtype APP in the presence of a water molecule was cap-
tured by GaMD simulations (Figure 8(a,d)). The water molecule that entered the presenilin active site was trapped
between two catalytic Asp residues that were 7 Å apart forming stable hydrogen bonds. A carbonyl oxygen of the scis-
sile amide bond present between Leu49 and Val50 residues of the APP substrate formed a hydrogen bond with
γ-secretase residue Asp257. During activation of γ-secretase, TM1, TM2, and TM8 helices of catalytic PS1 subunit
showed some flexibility while TM6a was observed interacting directly with the substrate. Free energy profiles revealed
four low-energy conformations of γ-secretase bound to wildtype APP, namely, “Inactive,” “Intermediate,” “Inhibited,”
and “Active” (Figure 8(d)). The active conformational state resembled the activation of the enzyme whereas the inactive
state correlated with the starting cryo-EM structure (Figures 8(a,b)). The active conformational state resembled the acti-
vation of the enzyme whereas the inactive state correlated with the starting cryo-EM structure. The inhibited conforma-
tional state closely resembled the γ-secretase structure in complex with DAPT inhibitor.162 The intermediate
conformational state resembled the transitional structure in between these states.
In addition to wildtype, GaMD simulations were performed on the APP mutants namely I45F, T48P, and M51F that
lead to FAD. The I45F and T48P system systems revealed faster activation compared to the wildtype system which were
in good agreement with the experimental mass spectrometry data of APP intracellular domain (AICD) proteolytic prod-
ucts that showed greater AICD50-99/AICD49-99 ratio in comparison to the wildtype. Four low-energy conformations,
similar to those observed in wildtype, were identified in the FAD mutant systems. Furthermore, the GaMD simulation
of M51F mutant shifted the ε cleavage between Thr48 and Leu49 (Figures 8(c,e)). The mass spectrometry analysis vali-
dated the AICD products formed as a result of proteolysis cleavage between Thr48 and Leu49 was observed higher.
In vitro assay showed significant high production of AICD in M51F system as compared to other systems. This was con-
sistent with the GaMD free energy landscape as the inhibited state was observed in wildtype, I45F, and T48P system
but not in M51F system. The experimental validations strongly correlated with the GaMD simulation model of
γ-secretase.
GaMD simulations of the wildtype and FAD mutant APPs of the different γ-secretase enzyme systems were ana-
lyzed with respect to their secondary structures. In particular, the M51F mutant APP shifted 4 Å downwards with
Thr48 and Leu49 residues flipping its side chain. The C-terminus of APP lost its β-sheet conformation as required for
local rearrangements during the shift in ε cleavage. The sub-pockets S10 , S20 , and S30 in the active site could also be
visualized in the wild-type and FAD mutant APPs bound to γ-secretase via GaMD simulations. Overall, the combina-
tion of GaMD simulations, mass spectrometry, and western blots enabled deep understanding of substrate processing
by γ-secretase and its activation.

3.5.7 | Structural dynamics of cytochrome P450

Cytochrome P450 3A4 (CYP34A) enzyme plays a crucial role in mammalian metabolic pathways including synthesis
and breakdown of fatty acids and hormones. It undergoes large conformational changes in the active site and other
WANG ET AL. 23 of 32

structural components of the enzyme. Redhair et al.92 performed GaMD simulations to understand the protein dynam-
ics and protein–ligand interactions induced by allosteric drug benzodiazepine midazolam (MDZ) in a lipid bilayer. The
GaMD simulations showed that F- and G-helical regions could be a possible allosteric site for MDZ drug, which were
further verified by hydrogen–deuterium exchange mass spectrometry. The GaMD simulations showed that the local
environment at the Phe-cluster comprising the region between the active site and the lipid bilayer was dynamic and
could be a possible allosteric site. Even in the presence of a ligand at the active site, the enzyme generated flexible allo-
steric site nearby. Overall, GaMD simulations provided the molecular basis of enzyme activity and allosteric drug inter-
action in cytochrome CYP34A.

3.6 | Carbohydrates

Here, we summarize recent GaMD studies of carbohydrates, fundamental components of cells that engage in energy
functions and form structural components. We review how GaMD has been applied to describe the dynamic interplay
of carbohydrates with DNA100 and proteins,58 and how the method has been used to facilitate the development of car-
bohydrate force field parameters.98

3.6.1 | The importance of carbohydrates conformation on DNA triplex

DNA triplexes are higher-order structural arrangements important for gene regulation and biotechnological applica-
tions. GaMD simulations were recently used to examine the impact of substituting deoxyribose sugars by con-
formationally locked sugars on the DNA triplex structure.100 Multiple GaMD simulations were performed on both 30 –50
and 50 –30 modified triplexes, as well as on an unmodified DNA triplex, which was used as a control. The simulations
revealed that the DNA triplexes, in which the deoxyribose was replaced by locked sugar, lost their structural integrity,
and disintegrated resembling the structure of a duplex. On the other hand, the control DNA triplex preserved the struc-
tural integrity during the simulations. As a notable observation, both modified triplexes changed conformation reaching
duplex structures containing a modified strand and a regular strand, while the third DNA strand was dissociated from
the complex. In-depth analysis of the trajectories indicated a significant reduction in the major groove width and dimin-
ished solvent accessible surface area in the modified triplexes, as compared to reference systems. On this basis, the
authors suggested that the newly introduced locked sugars impose a remarkable steric constraint, which alters the
DNA structure and results in the inefficient binding of the third DNA strand. Overall, the authors concluded that
knowledge of the structural changes induced by modified sugars could be leveraged for the design of new antisense oli-
gonucleotides, as well as to understand the role of modified oligonucleotides in anticancer therapy.

3.6.2 | Carbohydrate–protein interactions

Protein glycosylation is a post-translational modification that is involved in several cellular and biological processes.58
Glycosyltransferases (GTs) catalyze the transfer of a sugar moiety to acceptor amino acids, such as serine, threonine
(O-linked glycosylation), and asparagine (N-linked glycosylation). The glycosylation process also occurs in bacterial pro-
teins, where it plays a critical role in the immune response against pathogens. Some bacterial effectors leverage the gly-
cosylation process to suppress the nuclear factor NF-κB, which is central in regulating the immune response. The
nonlocus of enterocyte effacement effector protein B (NleB) has glycosyltransferase activity and inhibits NF-κB by
transferring N-acetyl glucosamine (GlcNAc). To further understand the glycosylation process, Park et al.58 focused on
the SseK1 and SseK2 effectors, which are orthologs of NleB from Salmonella typhimurium. The authors combined
X-ray crystallography, NMR, enzyme kinetics, GaMD simulations, and in vivo experiments. Structural evidence rev-
ealed a glycosyltransferase architecture displaying a helix–loop–helix (HLH) domain relevant to protein substrate recog-
nition and a catalytic core, which includes the conserved catalytic triad (His–Glu–Asn) critical for enzyme catalysis and
bacterial virulence. GaMD simulations showed large amplitude motions of the HLH domain, with significant differ-
ences in SseK1 and SseK2 that affect the HLH approach toward the substrate-binding pocket. Specifically, SseK1 was
considerably more flexible than SseK2 in the loop region connecting the HLH, in line with the increased specificity
toward the substrate that has been measured experimentally.58 The simulations also suggested a possible conformation
24 of 32 WANG ET AL.

of the catalytically competent active site, showing that the binding of GlcNAc properly orients the substrate, namely, an
arginine residue target of the glycosylation process, for the chemical reaction.

3.6.3 | Development of carbohydrate force field parameters with GaMD simulations

Heparin is a highly sulfated, linear polysaccharide belonging to the family of glycosaminoglycans. Endogenous heparin
critically regulates blood coagulation by interacting with the protein antithrombin (AT), through the pentasaccharide
domain responsible for the heparin activity.98,163 The capability of a heparin penta-saccharide to bind AT is determined
by the conformational dynamics of the sugar rings, and particularly by the conformation of the L-iduronic acid residue.
On this basis, idraparinux derivatives, which are nonglycosaminoglycan analogs of the heparin penta-saccharide, are
promising anticoagulant drug candidates. However, computational simulations of carbohydrates for drug discovery are
difficult due to their high flexibility. Moreover, the difficulty of computationally modeling idraparinux derivatives is
increased by the presence of sulfonato-methyl moieties, which are highly charged. Therefore, to attain a proper descrip-
tion of heparin and idraparinux derivatives, Balogh et al. assessed the performance of the GAFF1,164 GLYCAM06,165,166
and CHARMM167,168 force fields using GaMD simulations.98 This enabled enhancing the conformational landscape of
the pentasaccharide domain obtaining agreement with NMR experiments. The analysis of simulations and their com-
parison with NMR demonstrated that the CHARMM force field was better reproducing best the experimental data on
the ring conformations, producing also good agreement with the Nuclear Overhauser Effects (NOE) distances on L-
iduronic acid ring conformations. Therefore, the use of the CHARMM force field was proposed for the exhaustive and
comparative conformational analyses of idraparinux derivatives.

3.7 | Drug design

3.7.1 | Retrospective ensemble docking of allosteric modulators of A1AR

Virtual screening has been widely used for agonist/antagonist design targeting GPCRs.169 The success rate for virtual screen-
ing of GPCRs in the orthosteric pocket is mostly >20%, which is even higher than that of globular proteins.170 However, it is
rather challenging to apply virtual screening to identify allosteric modulators due to their low affinity compared with the
agonist/antagonist. In a recent study,102 we tested whether receptor structural ensembles obtained from GaMD simulations
could be used to increase docking performance of known PAMs using the A1AR as a model GPCR. Retrospective ensemble
docking calculations of PAMs to the A1AR combining GaMD simulations and Autodock171 were performed.
The GaMD simulations implemented in AMBER and NAMD were applied to generate receptor ensemble. The flexible
docking and rigid-body docking at different levels (short, medium, and long) were all evaluated. Docking scores corrected
by the GaMD reweighted free energy of the receptor structural cluster further improved the docking performances. The
calculated docking enrichment factors (EFs) and the area under the receiver operating characteristic curves (AUC) are
increased using ranking by the average binding energy (BEavg) in comparison with the minimum binding energy (BEmin).
Ensemble obtained from AMBER dual-boost GaMD simulations of the VCP171-bound ADO–A1AR–Gi complex
outperformed other ensembles for docking. Interactions between the PAM and receptor ECL2 in the VCP171-bound
ADO–A1AR–Gi complex might induce more suitable conformations for PAM binding, which were difficult to be sampled
in the simulations of PAM-free (apo) A1AR. Dual-boost GaMD with higher boost potential was observed to perform better
than the dihedral-boost GaMD for ensemble docking. Overall, flexible docking performed significantly better than the
rigid-body docking at different levels with AutoDock, suggesting that the flexibility of protein side chains is also important
in ensemble docking. In summary, the docking performance has been highly improved by combining GaMD simulations
and flexible docking, which effectively account for the flexibility of backbone and side-chain in receptors. Such an ensem-
ble docking protocol will greatly facilitate future PAM design of the A1AR and other GPCRs.

3.7.2 | Discovery of novel small-molecule calcium sensitizers for cardiac troponin C

Cardiac troponin C (TnC) is a calcium-dependent protein in the troponin complex responsible for the activation of mus-
cle contraction. Disorder of TnC may trigger heart diseases and then cause death. One of the current therapies172 is to
WANG ET AL. 25 of 32

design small molecules that can stabilize an open structure of the TnC and facilitate binding of the TnC switch peptide.
To identify potential small molecules for the treatment, Coldren et al.101 combined GaMD and high-throughput virtual
screening to predict binding conformations and affinities of small molecules in TnC. The simulations were compared
with experiments for the TnC protein structures in complex with calcium sensitivity modulators. Three hundred nano-
seconds GaMD simulations were performed on each system to obtain protein conformations. The simulation trajectory
snapshots were clustered into 10 most representative conformations based on an agglomerative hierarchical algorithm,
which were the structures used for virtual screening and docking studies. The work identified a number of novel com-
pounds that reduced the calcium dissociation rate and showed an overall calcium sensitization effect. One of the com-
pounds exhibited high binding affinity in TnC and was further verified by the stopped-flow kinetic experiment.

3.8 | Software

GaMD has been implemented in the common MD simulation packages, including both GPU and CPU versions of
AMBER,47,173 NAMD,51 and GENESIS60 (Table 1). All the above-mentioned GaMD algorithms have been incorporated
into the latest GPU version of AMBER 20.174

4 | C ON C L U S I ON S

Without the need to set predefined reaction coordinates or collective variables, GaMD can be advantageous for explor-
ing biomolecular conformational space and complex biomolecular interactions without a priori knowledge or con-
straints. Additionally, the boost potential in GaMD simulations exhibits a Gaussian distribution, allowing for accurate
reweighting of the simulations using cumulant expansion to the second order. Thus, GaMD is applicable to a wide
range of biological systems as described in this review.
Based on GaMD, novel approaches including rex-GaMD,59 GaREUS,60 and selective GaMD (LiGaMD and Pep-
GaMD)61,62 have been developed. The selective LiGaMD and Pep-GaMD algorithms appear to be more efficient and
easier to use as compared with other existing methods, including the cMD,175,176 replica exchange,177–179 and meta-
dynamics.179,180 For example, a binding event of drug Dasatinib to its binding site of Src Kinase was captured in a total
of 35 μs Anton cMD simulation.176 Repetitive binding and unbinding of an IDP peptide was captured in 200 μs Anton
cMD simulations at elevated temperature (400 K).175 Replica exchange algorithm needs to simultaneously simulate
many replicas to model ligand/peptide binding and dissociation.177–179 In comparison, LiGaMD and Pep-GaMD simula-
tions were able to capture multiple events of ligand/peptide binding and unbinding within microsecond simulation
time. These highly efficient simulations allowed us to accurately characterize the ligand/peptide binding thermodynam-
ics and kinetics.61,62
Finally, more efficient GaMD algorithms and enhanced sampling methods, in general, are still needed to character-
ize the thermodynamics and kinetics of important protein–protein/nucleic acid interactions, which could involve diffi-
cult simulations of both binding and dissociation of large biomolecular complexes. Moreover, structural dynamics in
systems of increasing sizes such as viruses and cells present grand challenges for computational modeling and
enhanced sampling simulations. Continued innovations in both supercomputing hardware and method developments
may help us to address these challenges in the future.

A C K N O WL E D G M E N T S
This work used supercomputing resources with allocation awards TG-MCB180049 (to Yinglong Miao) and TG-
MCB160059 (to Giulia Palermo) through the Extreme Science and Engineering Discovery Environment (XSEDE),
which is supported by National Science Foundation grant number ACI-1548562 and project M2874 through the
National Energy Research Scientific Computing Center (NERSC, to Yinglong Miao), which is a U.S. Department of
Energy Office of Science User Facility operated under Contract No. DE-AC02-05CH11231. It also used computational
resources provided by the Research Computing Cluster at the University of Kansas, the University of California Riv-
erside High-Performance Computing Center (HPCC), and the Triton Shared Computing Cluster (TSCC) of the San
Diego Supercomputing Center. This work was supported in part by the American Heart Association (Award
17SDG33370094), the National Institutes of Health (R01GM132572), and the startup funding in the College of Liberal
Arts and Sciences at the University of Kansas to Yinglong Miao. This work was also supported in part by the National
26 of 32 WANG ET AL.

Science Foundation under Grant No. CHE-1905374, by the National Institutes of Health (R01 EY027440) and startup
funding from the University of California Riverside to Giulia Palermo. We are grateful to the startup grant from
Wayne State University (to Yu-ming M. Huang).

A U T H O R C O N T R I B U T I O NS
Jinan Wang: Data curation; formal analysis; writing-original draft. Pablo Arantes: Data curation; formal analysis;
writing-original draft. Apurba Bhattarai: Formal analysis; writing-review & editing. Rohaine Hsu: Formal analysis;
writing-review & editing. Shristi Pawnikar: Formal analysis; writing-review & editing. Yu-ming Huang: Formal
analysis; writing-review & editing. Giulia Palermo: Conceptualization; funding acquisition; project administration;
resources; supervision; writing-original draft. Yinglong Miao: Conceptualization; funding acquisition; project adminis-
tration; resources; supervision; writing-original draft.

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

CONFLICT OF INTEREST
The authors have declared no conflict of interest for this article.

ORCID
Jinan Wang https://orcid.org/0000-0003-0162-212X
Pablo R. Arantes https://orcid.org/0000-0001-9707-8493
Apurba Bhattarai https://orcid.org/0000-0002-9904-8906
Rohaine V. Hsu https://orcid.org/0000-0001-6036-7208
Yu-ming M. Huang https://orcid.org/0000-0003-3257-6170
Giulia Palermo https://orcid.org/0000-0003-1404-8737
Yinglong Miao https://orcid.org/0000-0003-3714-1395

R EL ATE D WIR Es AR TI CL ES
Establishing the allosteric mechanism in CRISPR-Cas9

FURTHER READING
A website is available for GaMD at http://miao.compbio.ku.edu/GaMD/. GaMD has been implemented in the AMBER,
NAMD, and GENESIS software packages. User manuals and tutorials have been provided for using GaMD in these
packages through the GaMD website. For method updates and discussions such as bug reports and simulation ques-
tions, users can subscribe to the GaMD mailing list: https://sourceforge.net/projects/gamd/lists/gamd-discuss, and send
emails to gamd-discuss@lists.sourceforge.net.

R EF E RE N C E S
1. Henzler-Wildman K, Kern D. Dynamic personalities of proteins. Nature. 2007;450:964–72.
2. Hatoum-Aslan A, Marraffini LA. Impact of CRISPR immunity on the emergence and virulence of bacterial pathogens. Curr Opin
Microbiol. 2014;17:82–90.
3. Englander SW, Mayne L. The nature of protein folding pathways. Proc Natl Acad Sci USA. 2014;111:15873–80.
4. Ritter SL, Hall RA. Fine-tuning of GPCR activity by receptor-interacting proteins. Nat Rev Mol Cell Biol. 2009;10:819–30.
5. Onuchic JN, Luthey-Schulten Z, Wolynes PG. Theory of protein folding: the energy landscape perspective. Annu Rev Phys Chem. 1997;
48:545–600.
6. Deupi X, Kobilka BK. Energy landscapes as a tool to integrate GPCR structure, dynamics, and function. Phys Ther. 2010;25:293–303.
7. Karplus M, McCammon JA. Molecular dynamics simulations of biomolecules. Nat Struct Biol. 2002;9:646–52.
8. Hollingsworth SA, Dror RO. Molecular dynamics simulation for all. Neuron. 2018;99:1129–43.
9. Harvey MJ, Giupponi G, Fabritiis GD. ACEMD: accelerating biomolecular dynamics in the microsecond time scale. J Chem Theory
Comput. 2009;5:1632–9.
10. Johnston JM, Filizola M. Showcasing modern molecular dynamics simulations of membrane proteins through G protein-coupled recep-
tors. Curr Opin Struct Biol. 2011;21:552–8.
11. Shaw DE, Maragakis P, Lindorff-Larsen K, Piana S, Dror RO, Eastwood MP, et al. Atomic-level characterization of the structural
dynamics of proteins. Science. 2010;330:341–6.
WANG ET AL. 27 of 32

12. Lane TJ, Shukla D, Beauchamp KA, Pande VS. To milliseconds and beyond: challenges in the simulation of protein folding. Curr Opin
Struct Biol. 2013;23:58–65.
13. Vilardaga J-P, Bünemann M, Krasel C, Castro M, Lohse MJ. Measurement of the millisecond activation switch of G protein–coupled
receptors in living cells. Nat Biotechnol. 2003;21:807–12.
14. Miao Y, Ortoleva PJ. Viral structural transitions: an all-atom multiscale theory. J Chem Phys. 2006;125:214901.
15. Christen M, van Gunsteren WF. On searching in, sampling of, and dynamically moving through conformational space of biomolecular
systems: a review. J Comput Chem. 2008;29:157–66.
16. Spiwok V, Sucur Z, Hosek P. Enhanced sampling techniques in biomolecular simulations. Biotechnol Adv. 2015;33:1130–40.
17. Abrams C, Bussi G. Enhanced sampling in molecular dynamics using Metadynamics, replica-exchange, and temperature-acceleration.
Entropy. 2014;16:163–99.
18. Miao Y, McCammon JA. Unconstrained enhanced sampling for free energy calculations of biomolecules: a review. Mol Simul. 2016;42:
1046–55.
19. Torrie GM, Valleau JP. Nonphysical sampling distributions in Monte Carlo free-energy estimation: umbrella sampling. J Comput Phys.
1977;23:187–99.
20. Kumar S, Rosenberg JM, Bouzida D, Swendsen RH, Kollman PA. THE weighted histogram analysis method for free-energy calcula-
tions on biomolecules. I. THE method. J Comput Chem. 1992;13:1011–21.
21. Laio A, Gervasio FL. Metadynamics: a method to simulate rare events and reconstruct the free energy in biophysics, chemistry and
material science. Rep Prog Phys. 2008;71:126601.
22. Bešker N, Gervasio FL. Using metadynamics and path collective variables to study ligand binding and induced conformational transi-
tions. Computational drug discovery and design. Berlin: Springer; 2012. p. 501–13.
23. Darve E, Rodríguez-Gómez D, Pohorille A. Adaptive biasing force method for scalar and vector free energy calculations. J Chem Phys.
2008;128:144120.
24. Darve E, Wilson MA, Pohorille A. Calculating free energies using a scaled-force molecular dynamics algorithm. Mol Simul. 2002;28:
113–44.
25. Isralewitz B, Baudry J, Gullingsrud J, Kosztin D, Schulten K. Steered molecular dynamics investigations of protein function. J Mol
Graph Model. 2001;19:13–25.
26. Grubmuller H. Predicting slow structural transitions in macromolecular systems: conformational flooding. Phys Rev E Stat Phys
Plasmas Fluids Relat Interdiscip Topics. 1995;52:2893–906.
27. Bouvier B, Grubmuller H. A molecular dynamics study of slow base flipping in DNA using conformational flooding. Biophys J. 2007;93:
770–86.
28. Sugita Y, Okamoto Y. Replica-exchange molecular dynamics method for protein folding. Chem Phys Lett. 1999;314:141–51.
29. Okamoto Y. Generalized-ensemble algorithms: enhanced sampling techniques for Monte Carlo and molecular dynamics simulations.
J Mol Graph Model. 2004;22:425–39.
30. Hansmann UH. Parallel tempering algorithm for conformational studies of biological molecules. Chem Phys Lett. 1997;281:140–50.
31. Wu X, Wang S. Self-guided molecular dynamics simulation for efficient conformational search. J Phys Chem B. 1998;102:7238–50.
32. Wu X, Brooks BR. Self-guided Langevin dynamics simulation method. Chem Phys Lett. 2003;381:512–8.
33. Wu X, Brooks BR, Vanden-Eijnden E. Self-guided Langevin dynamics via generalized Langevin equation. J Comput Chem. 2016;37:
595–601.
34. Li H, Min D, Liu Y, Yang W. Essential energy space random walk via energy space metadynamics method to accelerate molecular
dynamics simulations. J Chem Phys. 2007;127:094101.
35. Zheng L, Yang W. Essential energy space random walks to accelerate molecular dynamics simulations: convergence improvements via
an adaptive-length self-healing strategy. J Chem Phys. 2008;129:014105.
36. Lv C, Zheng L, Yang W. Generalized essential energy space random walks to more effectively accelerate solute sampling in aqueous
environment. J Chem Phys. 2012;136:044103.
37. Hamelberg D, Mongan J, McCammon JA. Accelerated molecular dynamics: a promising and efficient simulation method for biomole-
cules. J Chem Phys. 2004;120:11919–29.
38. Voter AF. Hyperdynamics: accelerated molecular dynamics of infrequent events. Phys Rev Lett. 1997;78:3908–11.
39. Hamelberg D, de Oliveira CAF, McCammon JA. Sampling of slow diffusive conformational transitions with accelerated molecular
dynamics. J Chem Phys. 2007;127:10B614.
40. Shen T, Hamelberg D. A statistical analysis of the precision of reweighting-based simulations. J Chem Phys. 2008;129:034103.
41. Miao Y, Feixas F, Eun C, McCammon JA. Accelerated molecular dynamics simulations of protein folding. J Comput Chem. 2015;36:
1536–49.
42. Miao Y, Sinko W, Pierce L, Bucher D, Walker RC, McCammon JA. Improved reweighting of accelerated molecular dynamics simula-
tions for free energy calculation. J Chem Theory Comput. 2014;10:2677–89.
43. Jiang W, Thirman J, Jo S, Roux B. Reduced free energy perturbation/Hamiltonian replica exchange molecular dynamics method with
unbiased alchemical thermodynamic Axis. J Phys Chem B. 2018;122:9435–42.
44. Fajer M, Hamelberg D, McCammon JA. Replica-exchange accelerated molecular dynamics (REXAMD) applied to thermodynamic inte-
gration. J Chem Theory Comput. 2008;4:1565–9.
28 of 32 WANG ET AL.

45. Miao Y, Nichols SE, McCammon JA. Free energy landscape of G-protein coupled receptors, explored by accelerated molecular dynam-
ics. Phys Chem Chem Phys. 2014;16:6398–406.
46. Kappel K, Miao YL, McCammon JA. Accelerated molecular dynamics simulations of ligand binding to a muscarinic G-protein-coupled
receptor. Q Rev Biophys. 2015;48:479–87.
47. Miao Y, Caliman Alisha D, McCammon JA. Allosteric effects of sodium ion binding on activation of the M3 muscarinic G-protein-
coupled receptor. Biophys J. 2015;108:1796–806.
48. Miao Y, JA MC. Chapter six - Gaussian accelerated molecular dynamics: theory, implementation, and applications. In: Dixon DA, edi-
tor. Annual Report of Computational Chemistry. Volume 13. Amsterdam: Elsevier; 2017. p. 231–78.
49. Bhattarai A, Miao Y. Gaussian accelerated molecular dynamics for elucidation of drug pathways. Expert Opin Drug Discovery. 2018;13:
1055–65.
50. Miao Y, McCammon JA. Graded activation and free energy landscapes of a muscarinic G-protein–coupled receptor. Proc Natl Acad Sci
USA. 2016;113:12162–7.
51. Pang YT, Miao Y, Wang Y, McCammon JA. Gaussian accelerated molecular dynamics in NAMD. J Chem Theory Comput. 2017;
13:9–19.
52. Salawu EO. The impairment of TorsinA's binding to and interactions with its activator: an atomistic molecular dynamics study of pri-
mary dystonia. Front Mol Biosci. 2018;5:64.
53. Zhang J, Wang N, Miao Y, Hauser F, McCammon JA, Rappel W-J, et al. Identification of SLAC1 anion channel residues required for
CO2 bicarbonate sensing and regulation of stomatal movements. Proc Natl Acad Sci USA. 2018;115:11129–37.
54. Wang Y-T, Chan Y-H. Understanding the molecular basis of agonist/antagonist mechanism of human mu opioid receptor through
gaussian accelerated molecular dynamics method. Sci Rep. 2017;7:7828.
55. Liao JM, Wang YT. In silico studies of conformational dynamics of mu opioid receptor performed using gaussian accelerated molecular
dynamics. J Biomol Struct Dyn. 2019;37:166–77.
56. Chuang CH, Chiou SJ, Cheng TL, Wang YT. A molecular dynamics simulation study decodes the Zika virus NS5 methyltransferase
bound to SAH and RNA analogue. Sci Rep. 2018;8:6336.
57. Miao Y, Huang Y-mM, Walker RC, McCammon JA, Chang C-eA. Ligand binding pathways and conformational transitions of the HIV
protease. Biochemistry. 2018;57:1533–41.
58. Park JB, Kim YH, Yoo Y, Kim J, Jun SH, Cho JW, et al. Structural basis for arginine glycosylation of host substrates by bacterial effector
proteins. Nat Commun. 2018;9:4283.
59. Huang Y-mM, McCammon JA, Miao Y. Replica exchange Gaussian accelerated molecular dynamics: improved enhanced sampling
and free energy calculation. J Chem Theory Comput. 2018;14:1853–64.
60. Oshima H, Re S, Sugita Y. Replica-exchange umbrella sampling combined with Gaussian accelerated molecular dynamics for free-
energy calculation of biomolecules. J Chem Theory Comput. 2019;15:5199–208.
61. Miao Y, Bhattarai A, Wang J. Ligand Gaussian accelerated molecular dynamics (LiGaMD): characterization of ligand binding thermo-
dynamics and kinetics. J Chem Theory Comput. 2020;16:5526–47.
62. Wang J, Miao Y. Peptide Gaussian accelerated molecular dynamics (pep-GaMD): enhanced sampling and free energy and kinetics cal-
culations of peptide binding. J Chem Phys. 2020;153:154109.
63. Gao Y, Cao D, Pawnikar S, John KP, Ahn HM, Hill S, et al. Structure of the human respiratory syncytial virus M2-1 protein in complex
with a short positive-sense gene-end RNA. Structure. 2020;28:979–990.e974.
64. Wang J, Lan L, Wu X, Xu L, Miao Y. Mechanism of RNA recognition by a Musashi RNA-binding protein. bioRxiv 2020:
2020.2010.2030.362756.
65. Roy R, Mishra A, Poddar S, Nayak D, Kar P. Investigating the mechanism of recognition and structural dynamics of nucleoprotein-RNA
complex from Peste des petits ruminants virus via Gaussian accelerated molecular dynamics simulations. J Biomol Struct Dyn. 2020;1–14.
66. East KW, Newton JC, Morzan UN, Narkhede YB, Acharya A, Skeens E, et al. Allosteric motions of the CRISPR–Cas9 HNH nuclease
probed by NMR and molecular dynamics. J Am Chem Soc. 2020;142:1348–58.
67. Ricci CG, Chen JS, Miao Y, Jinek M, Doudna JA, McCammon JA, et al. Deciphering off-target effects in CRISPR-Cas9 through acceler-
ated molecular dynamics. ACS Cent Sci. 2019;5:651–62.
68. Wang J, Miao Y. Mechanistic insights into specific G protein interactions with adenosine receptors. J Phys Chem B. 2019;123:6462–73.
69. Miao Y, McCammon JA. Mechanism of the G-protein mimetic nanobody binding to a muscarinic G-protein-coupled receptor. Proc Natl
Acad Sci USA. 2018;115:3036–41.
70. Sibener LV, Fernandes RA, Kolawole EM, Carbone CB, Liu F, McAffee D, et al. Isolation of a structural mechanism for uncoupling T
cell receptor signaling from peptide-MHC binding. Cell. 2018;174:672–687.e627.
71. Lin W-W, Wang Y-J, Ko C-W, Cheng T-L, Wang Y-T. Cyclic peptide inhibitors of the Tsg101 UEV protein interactions refined through
global docking and Gaussian accelerated molecular dynamics simulations. Polymers. 2020;12:2235.
72. Petrizzelli F, Biagini T, Barbieri A, Parca L, Panzironi N, Castellana S, et al. Mechanisms of pathogenesis of missense mutations on the
KDM6A-H3 interaction in type 2 kabuki syndrome. Comput Struct Biotechnol J. 2020;18:2033–42.
73. Pawnikar S, Miao Y. Pathway and mechanism of drug binding to chemokine receptors revealed by accelerated molecular simulations.
Future Med Chem. 2020;12:1213–25.
74. Chen J, Wang W, Sun H, Pang L, Yin B. Mutation-mediated influences on binding of anaplastic lymphoma kinase to crizotinib decoded
by multiple replica Gaussian accelerated molecular dynamics. J Comput Aided Mol Des. 2020;34:1289–305.
WANG ET AL. 29 of 32

75. Paco L, Zarate-Perez F, Clouser AF, Atkins WM, Hackett JC. Dynamics and mechanism of binding of Androstenedione to membrane-
associated aromatase. Biochemistry. 2020;59:2999–3009.
76. Chen Y, Liu T, Xi Q, Jia W, Yin D, Wang X. A computational approach to the study of the binding mode of S1P1R agonists based on
the active-like receptor model. J Chem Inf Model. 2019;59:1624–33.
77. Tyagi C, Marik T, Vagvolgyi C, Kredics L, Otvos F. Accelerated molecular dynamics applied to the Peptaibol folding problem. Int J Mol
Sci. 2019;20:4268–87.
78. Peng Y, Cao S, Kiselar J, Xiao X, Du Z, Hsieh A, et al. A metastable contact and structural disorder in the estrogen receptor trans-
activation domain. Structure. 2019;27:229–240.e224.
79. Brown BP, Zhang Y-K, Westover D, Yan Y, Qiao H, Huang V, et al. On-target resistance to the mutant-selective EGFR inhibitor
Osimertinib can develop in an allele-specific manner dependent on the original EGFR-activating mutation. Clin Cancer Res. 2019;25:
3341–51.
80. Purslow JA, Nguyen TT, Egner TK, Dotas RR, Khatiwada B, Venditti V. Active site breathing of human Alkbh5 revealed by solution
NMR and accelerated molecular dynamics. Biophys J. 2018;115:1895–905.
81. Venkatramani A, Gravina Ricci C, Oldfield E, McCammon JA. Remarkable similarity in plasmodium falciparum and Plasmodium
vivax geranylgeranyl diphosphate synthase dynamics and its implication for antimalarial drug design. Chem Biol Drug Des. 2018;91:
1068–77.
82. Dodge GJ, Patel A, Jaremko KL, McCammon JA, Smith JL, Burkart MD. Structural and dynamical rationale for fatty acid unsaturation
in Escherichia coli. Proc Natl Acad Sci USA. 2019;116:6775–83.
83. Dujardin M, Madan V, Gandhi NS, Cantrelle FX, Launay H, Huvent I, et al. Cyclophilin A allows the allosteric regulation of a struc-
tural motif in the disordered domain 2 of NS5A and thereby fine-tunes HCV RNA replication. J Biol Chem. 2019;294:13171–85.
84. de Azevedo EC, Nascimento AS. Energy landscape of the domain movement in Staphylococcus aureus UDP-N-acetylglucosamine
2-epimerase. J Struct Biol. 2019;207:158–68.
85. Han J, Shi X, Du Y, Shi F, Zhang B, Zheng Z, et al. Schisandrin C targets Keap1 and attenuates oxidative stress by activating Nrf2 path-
way in Ang II-challenged vascular endothelium. Phytother Res. 2019;33:779–90.
86. Wu X, Wang Y, Jia R, Fang F, Liu Y, Cui W. Computational and biological investigation of the soybean lecithin-gallic acid complex for
ameliorating alcoholic liver disease in mice with iron overload. Food Funct. 2019;10:5203–14.
87. Koh A, Gibbon MJ, Van der Kamp MW, Pudney CR, Gebhard S. Conformation control of the histidine kinase BceS of Bacillus subtilis
by its cognate ABC-transporter facilitates need-based activation of antibiotic resistance. Mol Microbiol. 2020;115:157–74.
88. Moffett AS, Shukla D. Structural consequences of multisite phosphorylation in the BAK1 kinase domain. Biophys J. 2020;118:698–707.
89. Philpott JM, Narasimamurthy R, Ricci CG, Freeberg AM, Hunt SR, Yee LE, et al. Casein kinase 1 dynamics underlie substrate selectiv-
ity and the PER2 circadian phosphoswitch. Elife. 2020;9:e52343.
90. Sztain T, Amaro R, McCammon JA. Elucidation of cryptic and allosteric pockets within the SARS-CoV-2 protease. bioRxiv 2020:
2020.2007.2023.218784.
91. Zhang Q, Tan S, Xiao T, Liu H, Shah SJA, Liu H. Probing the molecular mechanism of rifampin resistance caused by the point muta-
tions S456L and D441V on Mycobacterium tuberculosis RNA polymerase through Gaussian accelerated molecular dynamics simulation.
Antimicrob Agents Chemother. 2020;64:e02476-02419.
92. Redhair M, Hackett JC, Pelletier RD, Atkins WM. Dynamics and location of the allosteric midazolam site in cytochrome P4503A4 in
lipid Nanodiscs. Biochemistry. 2020;59:766–79.
93. An X, Bai Q, Bing Z, Zhou S, Shi D, Liu H, et al. How does agonist and antagonist binding Lead to different conformational ensemble
Equilibria of the kappa-opioid receptor: insight from long-time Gaussian accelerated molecular dynamics simulation. ACS Chem
Nerosci. 2019;10:1575–84.
94. Conrad M, Soldner CA, Miao Y, Sticht H. Agonist binding and G protein coupling in histamine H2 receptor: a molecular dynamics
study. Int J Mol Sci. 2020;21:6693–712.
95. Bhattarai A, Wang J, Miao Y. G-protein-coupled receptor-membrane interactions depend on the receptor activation state. J Comput
Chem. 2020;41:460–71.
96. Zhao Y, Ung PM, Zahoranszky-Kohalmi G, Zakharov AV, Martinez NJ, Simeonov A, et al. Identification of a G-protein-independent
activator of GIRK channels. Cell Rep. 2020;31:107770.
97. Bhattarai A, Devkota S, Bhattarai S, Wolfe MS, Miao Y. Mechanisms of gamma-Secretase activation and substrate processing. ACS Cent
Sci. 2020;6:969–83.
98. Balogh G, Gyöngyösi T, Timári I, Herczeg M, Borbás A, Fehér K, et al. Comparison of carbohydrate force fields using Gaussian acceler-
ated molecular dynamics simulations and development of force field parameters for heparin-analogue Pentasaccharides. J Chem Inf
Model. 2019;59:4855–67.
99. Balogh G, Komaromi I, Bereczky Z. The mechanism of high affinity pentasaccharide binding to antithrombin, insights from Gaussian
accelerated molecular dynamics simulations. J Biomol Struct Dyn. 2020;38:4718–32.
100. Pant P, Fisher M. DNA triplex with conformationally locked sugar disintegrates to duplex: insights from molecular simulations. Bio-
chem Biophys Res Commun. 2020;532:662–7.
101. Coldren WH, Tikunova SB, Davis JP, Lindert S. Discovery of novel small-molecule calcium sensitizers for cardiac troponin C: a com-
bined virtual and experimental screening approach. J Chem Inf Model. 2020;60:3648–61.
30 of 32 WANG ET AL.

102. Bhattarai A, Wang J, Miao Y. Retrospective ensemble docking of allosteric modulators in an adenosine G-protein-coupled receptor. Bio-
chim Biophys Acta Gen Subj. 2020;1864:129615.
103. Hummer G. Fast-growth thermodynamic integration: error and efficiency analysis. J Chem Phys. 2001;114:7330–7.
104. Eastwood MP, Hardin C, Luthey-Schulten Z, Wolynes PG. Statistical mechanical refinement of protein structure prediction schemes:
cumulant expansion approach. J Chem Phys. 2002;117:4602–15.
105. Vanommeslaeghe K, MacKerell AD Jr. CHARMM additive and polarizable force fields for biophysics and computer-aided drug design.
Biochim Biophys Acta Gen Subj. 2015;1850:861–71.
106. Duan Y, Wu C, Chowdhury S, Lee MC, Xiong G, Zhang W, et al. A point-charge force field for molecular mechanics simulations of pro-
teins based on condensed-phase quantum mechanical calculations. J Comput Chem. 2003;24:1999–2012.
107. Miao Y. Acceleration of biomolecular kinetics in Gaussian accelerated molecular dynamics. J Chem Phys. 2018;149:072308.
108. Doudna JA, Charpentier E. Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science. 2014;346:1258096.
109. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E. A programmable dual-RNA-guided DNA endonuclease in adap-
tive bacterial immunity. Science. 2012;337:816–21.
110. Zhu X, Clarke R, Puppala AK, Chittori S, Merk A, Merrill BJ, et al. Cryo-EM structures reveal coordinated domain motions that govern
DNA cleavage by Cas9. Nat Struct Mol Biol. 2019;26:679–85.
111. Jiang F, Taylor DW, Chen JS, Kornfeld JE, Zhou K, Thompson AJ, et al. Structures of a CRISPR-Cas9 R-loop complex primed for DNA
cleavage. Science. 2016;351:867–71.
112. Jinek M, Jiang F, Taylor DW, Sternberg SH, Kaya E, Ma E, et al. Structures of Cas9 endonucleases reveal RNA-mediated conforma-
tional activation. Science. 2014;343:1247997.
113. Jiang F, Zhou K, Ma L, Gressel S, Doudna JA. Structural biology. A Cas9-guide RNA complex preorganized for target DNA recognition.
Science. 2015;348:1477–81.
114. Palermo G, Miao Y, Walker RC, Jinek M, McCammon JA. CRISPR-Cas9 conformational activation as elucidated from enhanced molec-
ular simulations. Proc Natl Acad Sci USA. 2017;114:7260–5.
115. Nishimasu H, Ran FA, Hsu PD, Konermann S, Shehata SI, Dohmae N, et al. Crystal structure of Cas9 in complex with guide RNA and
target DNA. Cell. 2014;156:935–49.
116. Palermo G, Miao Y, Walker RC, Jinek M, McCammon JA. Striking plasticity of CRISPR-Cas9 and key role of non-target DNA, as rev-
ealed by molecular simulations. ACS Cent Sci. 2016;2:756–63.
117. Dagdas YS, Chen JS, Sternberg SH, Doudna JA, Yildiz A. A conformational checkpoint between DNA binding and cleavage by
CRISPR-Cas9. Sci Adv. 2017;3:eaao0027.
118. Sternberg SH, LaFrance B, Kaplan M, Doudna JA. Conformational control of DNA target cleavage by CRISPR-Cas9. Nature. 2015;527:110–3.
119. Casalino L, Nierzwicki Ł, Jinek M, Palermo G. Catalytic mechanism of non-target DNA cleavage in CRISPR-Cas9 revealed by Ab-initio
molecular dynamics. ACS Catal. 2020;10:13596–605. https://doi.org/10.1021/acscatal.0c03566.
120. Palermo G. Structure and dynamics of the CRISPR–Cas9 catalytic complex. J Chem Inf Model. 2019;59:2394–406.
121. Shaw DE, Grossman J, Bank JA, Batson B, Butts JA, Chao JC, Deneroff MM, Dror RO, Even A, Fenton CH. Anton 2: raising the bar
for performance and programmability in a special-purpose molecular dynamics supercomputer. In: SC'14: Proceedings of the Interna-
tional Conference for High Performance Computing, Networking, Storage and Analysis. IEEE; 2014.
122. Palermo G, Chen JS, Ricci CG, Rivalta I, Jinek M, Batista VS, et al. Key role of the REC lobe during CRISPR–Cas9 activation by
‘sensing’,‘regulating’, and ‘locking'the catalytic HNH domain. Q Rev Biophys. 2018;51:e9.
123. Yang M, Peng S, Sun R, Lin J, Wang N, Chen C. The conformational dynamics of Cas9 governing DNA cleavage are revealed by single-
molecule FRET. Cell Rep. 2018;22:372–82.
124. Chen JS, Dagdas YS, Kleinstiver BP, Welch MM, Sousa AA, Harrington LB, et al. Enhanced proofreading governs CRISPR–Cas9
targeting accuracy. Nature. 2017;550:407–10.
125. Mitchell BP, Hsu RV, Medrano MA, Zewde NT, Narkhede YB, Palermo G. Spontaneous embedding of DNA mismatches within the
RNA: DNA hybrid of CRISPR-Cas9. Front Mol Biosci. 2020;7:39.
126. Schmid-Burgk JL, Gao L, Li D, Gardner Z, Strecker J, Lash B, et al. Highly parallel profiling of Cas9 variant specificity. Mol Cell. 2020;
78:794–800.e798.
127. Slaymaker IM, Gao L, Zetsche B, Scott DA, Yan WX, Zhang F. Rationally engineered Cas9 nucleases with improved specificity. Science.
2016;351:84–8.
128. Nierzwicki Ł, Arantes PR, Saha A, Palermo G. Establishing the allosteric mechanism in CRISPR-Cas9. WIREs Comput Mol Sci. 2020;
e1503.
129. Palermo G, Ricci CG, Fernando A, Basak R, Jinek M, Rivalta I, et al. Protospacer adjacent motif-induced Allostery activates CRISPR-
Cas9. J Am Chem Soc. 2017;139:16028–31.
130. Sternberg SH, Redding S, Jinek M, Greene EC, Doudna JA. DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature.
2014;507:62–7.
131. Sethi A, Eargle J, Black AA, Luthey-Schulten Z. Dynamical networks in tRNA:protein complexes. Proc Natl Acad Sci USA. 2009;106:
6620–5.
132. Ahrens VM, Bellmann-Sickert K, Beck-Sickinger AG. Peptides and peptide conjugates: therapeutics on the upward path. Future Med
Chem. 2012;4:1567–86.
133. Fosgerau K, Hoffmann T. Peptide therapeutics: current status and future directions. Drug Discov Today. 2015;20:122–8.
WANG ET AL. 31 of 32

134. Kahler U, Fuchs JE, Goettig P, Liedl KR. An unexpected switch in peptide binding mode: from simulation to substrate specificity.
J Biomol Struct Dyn. 2018;36:4072–84.
135. Andreani J, Guerois R. Evolution of protein interactions: from interactomes to interfaces. Arch Biochem Biophys. 2014;554:65–75.
136. Arkin MR, Wells JA. Small-molecule inhibitors of protein-protein interactions: progressing towards the dream. Nat Rev Drug Discov.
2004;3:301–17.
137. Wang J, Alekseenko A, Kozakov D, Miao Y. Improved modeling of peptide-protein binding through global docking and accelerated
molecular dynamics simulations. Front Mol Biosci. 2019;6:112.
138. Porter KA, Xia B, Beglov D, Bohnuud T, Alam N, Schueler-Furman O, et al. ClusPro PeptiDock: efficient global docking of peptide rec-
ognition motifs using FFT. Bioinformatics. 2017;33:3299–301.
139. Wang Y-T, Cheng T-L. Computational modeling of cyclic peptide inhibitor–MDM2/MDMX binding through global docking and Gauss-
ian accelerated molecular dynamics simulations. J Biomol Struct Dyn. 2020;1–10.
140. Ahmad M, Helms V. How do proteins associate? A lesson from SH3 domain. Chem Cent J. 2009;3:O22.
141. Ball LJ, Kuhne R, Schneider-Mergener J, Oschkinat H. Recognition of proline-rich motifs by protein-protein-interaction domains.
Angew Chem Int Ed Engl. 2005;44:2852–69.
142. Wang L, Friesner RA, Berne BJ. Replica exchange with solute scaling: a more efficient version of replica exchange with solute temper-
ing (REST2). J Phys Chem B. 2011;115:9431–8.
143. Hussein HA, Borrel A, Geneix C, Petitjean M, Regad L, Camproux A-C. PockDrug-server: a new web server for predicting pocket
druggability on holo and apo proteins. Nucleic Acids Res. 2015;43:W436–42.
144. Kruse AC, Ring AM, Manglik A, Hu J, Hu K, Eitel K, et al. Activation and allosteric modulation of a muscarinic acetylcholine receptor.
Nature. 2013;504:101–6.
145. Draper-Joyce CJ, Khoshouei M, Thal DM, Liang YL, Nguyen ATN, Furness SGB, et al. Structure of the adenosine-bound human adeno-
sine A1 receptor-Gi complex. Nature. 2018;558:559–63.
146. Garcia-Nafria J, Lee Y, Bai X, Carpenter B, Tate CG. Cryo-EM structure of the adenosine A2A receptor coupled to an engineered het-
erotrimeric G protein. Elife. 2018;7:e35946.
147. Glukhova A, Thal DM, Nguyen AT, Vecchio EA, Jorg M, Scammells PJ, et al. Structure of the adenosine A1 receptor reveals the basis
for subtype selectivity. Cell. 2017;168:867–877.e813.
148. Gao ZG, Inoue A, Jacobson KA. On the G protein-coupling selectivity of the native A2B adenosine receptor. Biochem Pharmacol. 2018;
151:201–13.
149. Cordeaux Y, Ijzerman AP, Hill SJ. Coupling of the human A1 adenosine receptor to different heterotrimeric G proteins: evidence for
agonist-specific G protein activation. Br J Pharmacol. 2004;143:705–14.
150. Cordeaux Y, Briddon SJ, Megson AE, McDonnell J, Dickenson JM, Hill SJ. Influence of receptor number on functional responses
elicited by agonists acting at the human adenosine A(1) receptor: evidence for signaling pathway-dependent changes in agonist potency
and relative intrinsic activity. Mol Pharmacol. 2000;58:1075–84.
151. Cheng RKY, Segala E, Robertson N, Deflorian F, Dore AS, Errey JC, et al. Structures of human A1 and A2A adenosine receptors with
Xanthines reveal determinants of selectivity. Structure. 2017;25:1275–1285.e1274.
152. Perly B, Smith IC, Jarrell HC. Acyl chain dynamics of phosphatidylethanolamines containing oleic acid and dihydrosterculic acid: deu-
teron NMR relaxation studies. Biochemistry. 1985;24:4659–65.
153. Shneyvays V, Leshem D, Zinman T, Mamedova LK, Jacobson KA, Shainberg A. Role of adenosine A1 and A3 receptors in regu-
lation of cardiomyocyte homeostasis after mitochondrial respiratory chain injury. Am J Physiol Heart Circ Physiol. 2005;288:
H2792–801.
154. Schutte F, Burgdorf C, Richardt G, Kurz T. Adenosine A1 receptor-mediated inhibition of myocardial norepinephrine release involves
neither phospholipase C nor protein kinase C but does involve adenylyl cyclase. Can J Physiol Pharmacol. 2006;84:573–7.
155. Liang BT. Protein kinase C-mediated preconditioning of cardiac myocytes: role of adenosine receptor and KATP channel. Am J Physiol
Heart Circ Physiol. 1997;273:H847–53.
156. Kiesman WF, Elzein E, Zablocki J. A 1 adenosine receptor antagonists, agonists, and allosteric enhancers. Adenosine receptors in health
and disease. Berlin: Springer; 2009. p. 25–58.
157. Romagnoli R, G Baraldi P, A Tabrizi M, Gessi S, A Borea P, Merighi S. Allosteric enhancers of A1 adenosine receptors: state of the art
and new horizons for drug development. Curr Med Chem. 2010;17:3488–502.
158. Miao Y, Bhattarai A, Nguyen ATN, Christopoulos A, May LT. Structural basis for binding of allosteric drug leads in the adenosine A1
receptor. Sci Rep. 2018;8:16836.
159. Nguyen AT, Vecchio EA, Thomas T, Nguyen TD, Aurelio L, Scammells PJ, et al. Role of the second extracellular loop of the adenosine
A1 receptor on allosteric modulator binding, signaling, and Cooperativity. Mol Pharmacol. 2016;90:715–25.
160. Peeters MC, Wisse LE, Dinaj A, Vroling B, Vriend G, Ijzerman AP. The role of the second and third extracellular loops of the adenosine
A1 receptor in activation and allosteric modulation. Biochem Pharmacol. 2012;84:76–87.
161. Zhou R, Yang G, Guo X, Zhou Q, Lei J, Shi Y. Recognition of the amyloid precursor protein by human gamma-secretase. Science. 2019;
363:eaaw0930.
162. Bai XC, Rajendra E, Yang G, Shi Y, Scheres SH. Sampling the conformational space of the catalytic subunit of human gamma-secretase.
Elife. 2015;4:e11182.
32 of 32 WANG ET AL.

163. Cui JY, Zhang F, Nierzwicki L, Palermo G, Linhardt RJ, Lisi GP. Mapping the structural and dynamic determinants of pH-sensitive
heparin binding to granulocyte macrophage Colony stimulating factor. Biochemistry. 2020;59:3541–53.
164. Wang J, Wolf RM, Caldwell JW, Kollman PA, Case DA. Development and testing of a general amber force field. J Comput Chem. 2004;
25:1157–74.
165. Singh A, Tessier MB, Pederson K, Wang X, Venot AP, Boons GJ, et al. Extension and validation of the GLYCAM force field parameters
for modeling glycosaminoglycans. Can J Chem. 2016;94:927–35.
166. Kirschner KN, Yongye AB, Tschampel SM, Gonzalez-Outeirino J, Daniels CR, Foley BL, et al. GLYCAM06: a generalizable biomolecu-
lar force field. Carbohydrates. J Comput Chem. 2008;29:622–55.
167. Mallajosyula SS, Guvench O, Hatcher E, Mackerell AD Jr. CHARMM additive all-atom force field for phosphate and sulfate linked to
carbohydrates. J Chem Theory Comput. 2012;8:759–76.
168. Guvench O, Mallajosyula SS, Raman EP, Hatcher E, Vanommeslaeghe K, Foster TJ, et al. CHARMM additive all-atom force field for
carbohydrate derivatives and its utility in polysaccharide and carbohydrate-protein modeling. J Chem Theory Comput. 2011;7:3162–80.
169. Wang J, Bhattarai A, Ahmad WI, Farnan TS, John KP, Miao Y. Chapter 15—computer-aided GPCR drug discovery. In: Jastrzebska B,
Park PSH, editors. GPCRs. New York: Academic Press; 2020. p. 283–93.
170. Shoichet BK, Kobilka BK. Structure-based drug screening for G-protein-coupled receptors. Trends Pharmacol Sci. 2012;33:268–72.
171. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, et al. AutoDock4 and AutoDockTools4: automated docking
with selective receptor flexibility. J Comput Chem. 2009;30:2785–91.
172. Remme WJ, Swedberg K, Task Force for the Diagnosis and Treatment of Chronic Heart Failure, European Society of Cardiology.
Guidelines for the diagnosis and treatment of chronic heart failure. Eur Heart J. 2001;22:1527–60.
173. D.A. Case K. Belfon, I.Y. Ben-Shalom, S.R. Brozell, D.S. Cerutti, T.E. Cheatham, III, T.A. Darden, R.E. Duke, T.J. Giese, H. Gohlke, A.
W. Goetz, D. Greene, N. Homeyer, S. Izadi, A. Kovalenko, T.S. Lee, S. LeGrand, P. Li, C. Lin, J. Liu, T. Luchko, R. Luo, D.
Mermelstein, K.M. Merz, G. Monard, H. Nguyen, I. Omelyan, A. Onufriev, F. Pan, R. Qi, D.R. Roe, A. Roitberg, C. Sagui, C.L.
Simmerling, W.M. Botello-Smith, J. Swails, R.C. Walker, J. Wang, R.M. Wolf, X. Wu, L. Xiao, D.M. York and P.A. Kollman (2020),
AMBER 2020, University of California, San Francisco.
174. Case DA, Belfon K, Ben-Shalom IY, Brozell SR, Cerutti DS, T. E. Cheatham I, Cruzeiro VWD, Darden TA, Duke RE, Giambasu G,
et al. AMBER 20, University of California, San Francisco. 2020.
175. Robustelli P, Piana S, Shaw DE. Mechanism of coupled folding-upon-binding of an intrinsically disordered protein. J Am Chem Soc.
2020;142:11092–101.
176. Shan Y, Kim ET, Eastwood MP, Dror RO, Seeliger MA, Shaw DE. How does a drug molecule find its target binding site? J Am Chem
Soc. 2011;133:9181–3.
177. Morrone JA, Perez A, MacCallum J, Dill KA. Computed binding of peptides to proteins with MELD-accelerated molecular dynamics.
J Chem Theory Comput. 2017;13:870–6.
178. Morrone JA, Perez A, Deng Q, Ha SN, Holloway MK, Sawyer TK, et al. Molecular simulations identify binding poses and approximate
affinities of stapled α-helical peptides to MDM2 and MDMX. J Chem Theory Comput. 2017;13:863–9.
179. Zou R, Zhou Y, Wang Y, Kuang G, Ågren H, Wu J, et al. Free energy profile and kinetics of coupled folding and binding of the intrinsi-
cally disordered protein p53 with MDM2. J Chem Inf Model. 2020;60:1551–8.
180. Tiwary P, Limongelli V, Salvalaglio M, Parrinello M. Kinetics of protein–ligand unbinding: predicting pathways, rates, and rate-limiting
steps. Proc Natl Acad Sci USA. 2015;112:E386–91.

How to cite this article: Wang J, Arantes PR, Bhattarai A, et al. Gaussian accelerated molecular dynamics:
Principles and applications. WIREs Comput Mol Sci. 2021;e1521. https://doi.org/10.1002/wcms.1521