Many
chemical,
physical,
and
biological
processes
occur
in
complex
environments
span
multiple
scales
space
time.
Combined
(hybrid
or
integrated)
quantum-mechanics/molecular-mechanics
(QM/MM)
is
one
family
of
multiscale
algorithms
for
computer
modeling
these
processes.
First
introduced
1976,
along
with
critical
developments
over
the
following
decades,
QM/MM
now
a
popular
powerful
tool
that
helps
scientists
simultaneously
capture
fundamental
atomistic
details
overall
big
picture
Instead
providing
comprehensive
survey
their
applications,
Methods
focuses
on
explaining
key
concepts
methodology
how
to
interpret
results
applications.
The
author
frequently
uses
small
"toy"
model
systems
illustrate
principles,
which
can
be
easily
generalized
large
systems.
Unavoidably,
formal
proofs
many
technical
are
left
out,
readers
referred
relevant
literature.
researchers
enter
field
good
starting
position,
empowering
them
ask
right
questions
applications
select
suitable
address
them.
Pharmaceutics,
Год журнала:
2025,
Номер
17(3), С. 315 - 315
Опубликована: Март 1, 2025
Chinese
materia
medica
(CMM)
refers
to
the
medicinal
substances
used
in
traditional
medicine.
In
recent
years,
CMM
has
become
globally
prevalent,
and
scientific
research
on
increasingly
garnered
attention.
Computer-aided
drug
design
(CADD)
been
employed
Western
medicine
for
many
contributing
significantly
its
progress.
However,
role
of
CADD
not
systematically
reviewed.
This
review
briefly
introduces
methods
from
perspectives
computational
chemistry
(including
quantum
chemistry,
molecular
mechanics,
mechanics/molecular
mechanics)
informatics
cheminformatics,
bioinformatics,
data
mining).
Then,
it
provides
an
exhaustive
discussion
applications
these
through
rich
cases.
Finally,
outlines
advantages
challenges
research.
conclusion,
despite
current
challenges,
still
offers
unique
over
experiments.
With
development
industry
computer
science,
especially
driven
by
artificial
intelligence,
is
poised
play
pivotal
advancing
Journal of Chemical Theory and Computation,
Год журнала:
2025,
Номер
unknown
Опубликована: Март 21, 2025
Creating
analytic
representations
of
multiple
potential
energy
surfaces
for
modeling
electronically
nonadiabatic
processes
is
a
major
challenge
being
addressed
in
various
ways
by
the
chemical
dynamics
community.
In
this
work,
we
introduce
new
method
that
can
achieve
convenient
learning
(PESs)
and
their
gradients
(negatives
forces)
polyatomic
system.
This
method,
called
compatibilization
deep
neural
network
(CDNN),
demonstrated
to
be
accurate
and,
even
more
importantly,
automatic.
The
only
required
input
database
with
geometries
energies.
produces
matrix,
compatible
matrix
(CPEM),
may
interpreted
as
electronic
Hamiltonian
an
implicit
basis,
adiabatic
are
obtained
diagonalization
automatic
differentiation.
We
show
CPEM,
which
neither
nor
necessarily
diabatic,
discovered
automatically
during
procedure
special
design
CDNN
architecture.
believe
will
very
useful
practice
coupled
PESs
systems
because
it
fully
The Journal of Chemical Physics,
Год журнала:
2024,
Номер
161(3)
Опубликована: Июль 15, 2024
PyDFT-QMMM
is
a
Python-based
package
for
performing
hybrid
quantum
mechanics/molecular
mechanics
(QM/MM)
simulations
at
the
density
functional
level
of
theory.
The
program
designed
to
treat
short-range
and
long-range
interactions
through
user-specified
combinations
electrostatic
mechanical
embedding
procedures
within
periodic
simulation
domains,
providing
necessary
interfaces
external
chemistry
molecular
dynamics
software.
To
enable
direct
electrostatics
in
systems,
we
have
derived
implemented
force
terms
our
previously
described
QM/MM/PME
approach
[Pederson
McDaniel,
J.
Chem.
Phys.
156,
174105
(2022)].
Communication
with
software
packages
Psi4
OpenMM
facilitated
Python
application
programming
(APIs).
core
library
contains
basic
utilities
running
QM/MM
simulations,
plug-in
entry-points
are
provided
users
implement
custom
energy/force
calculation
integration
routines,
an
extensible
architecture.
user
interacts
primarily
its
API,
allowing
complex
workflow
development
scripting,
example,
interfacing
PLUMED
free
energy
simulations.
We
provide
benchmarks
forces
conservation
alternative
approaches.
further
demonstrate
simple
example
use
case
water
solute
solvent
system,
which
radial
distribution
functions
computed
from
100
ps
simulations;
this
highlight
how
solvation
structure
sensitive
different
basis-set
choices
due
under-
or
over-polarization
QM
molecule’s
electron
density.
The Journal of Physical Chemistry A,
Год журнала:
2024,
Номер
unknown
Опубликована: Дек. 4, 2024
Despite
the
success
and
widespread
use
of
QM/MM
methods
in
modeling
(bio)chemically
important
processes,
their
accuracy
is
still
not
well
understood.
A
key
reason
because
these
are
ultimately
approximations
to
direct
QM
calculations
very
large
systems,
which
impractical
perform
most
cases.
We
highlight
recent
progress
toward
development
realistic
model
systems
where
it
possible
obtain
full
reference
data
directly
systematically
evaluate
effectiveness
different
generation
schemes.
These
highly
flexible
can
be
tailored
probe
sensitivity
a
reaction
types
simulation
parameters
such
as
pairing
MM
potentials,
region
size,
composition.
It
envisaged
that
this
strategy
could
used
validate
schemes
spur
more
robust
models
future.
The Journal of Chemical Physics,
Год журнала:
2024,
Номер
160(23)
Опубликована: Июнь 17, 2024
Understanding
the
nonadiabatic
dynamics
of
complex
systems
is
a
challenging
task
in
computational
photochemistry.
Herein,
we
present
an
efficient
and
user-friendly
quantum
mechanics/molecular
mechanics
(QM/MM)
interface
to
run
on-the-fly
dynamics.
Currently,
this
consists
independent
set
codes
designed
for
general-purpose
use.
demonstrate
ability
feasibility
QM/MM
by
integrating
it
with
our
long-term
developed
JADE
package.
Tailored
handle
processes
various
systems,
especially
condensed
phases
protein
environments,
delve
into
theories,
implementations,
applications
The
approach
established
within
framework
additive
scheme,
employing
electrostatic
embedding,
link-atom
inclusion,
charge-redistribution
schemes
treat
boundary.
Trajectory
surface-hopping
are
facilitated
using
fewest
switches
algorithm,
encompassing
classical
treatments
nuclear
electronic
motions,
respectively.
Finally,
report
simulations
two
typical
systems:
azomethane
water
retinal
chromophore
PSB3
environment.
Our
results
not
only
illustrate
power
program
but
also
reveal
important
roles
environmental
factors
processes.
Current Opinion in Structural Biology,
Год журнала:
2024,
Номер
86, С. 102821 - 102821
Опубликована: Апрель 29, 2024
The
complexity
of
biological
systems
and
processes,
spanning
molecular
to
macroscopic
scales,
necessitates
the
use
multiscale
simulations
get
a
comprehensive
understanding.
Quantum
mechanics/molecular
mechanics
(QM/MM)
dynamics
(MD)
are
crucial
for
capturing
processes
beyond
reach
classical
MD
simulations.
advent
exascale
computing
offers
unprecedented
opportunities
scientific
exploration,
not
least
within
life
sciences,
where
essential
unravel
intricate
mechanisms
underlying
processes.
However,
leveraging
immense
computational
power
requires
innovative
algorithms
software
designs.
In
this
context,
we
discuss
current
status
future
prospects
biomolecular
on
supercomputers
with
focus
QM/MM
MD.
We
highlight
our
own
efforts
in
developing
versatile
high-performance
simulation
framework
aim
efficient
utilization
state-of-the-art
supercomputers.
showcase
its
application
uncovering
complex
potential
computing.
The Journal of Chemical Physics,
Год журнала:
2024,
Номер
160(21)
Опубликована: Июнь 4, 2024
The
MolSSI
Driver
Interface
(MDI)
Project
is
an
effort
to
simplify
and
standardize
the
process
of
enabling
tight
interoperability
between
independently
developed
code
bases
supported
by
numerous
software
packages
across
domain
chemical
physics.
It
enables
a
wide
variety
use
cases,
including
quantum
mechanics/molecular
mechanics,
advanced
sampling,
path
integral
molecular
dynamics,
machine
learning,
ab
initio
etc.
We
describe
two
major
developments
within
MDI
that
provide
novel
solutions
key
challenges.
first
these
development
Plugin
System,
which
allows
MDI-supporting
libraries
be
used
as
highly
modular
plugins,
with
enforcing
standardized
application
programming
interface
plugins.
Codes
can
plugins
without
linking
against
them
during
their
build
process,
end-users
select
plugin(s)
they
wish
at
runtime.
System
features
sophisticated
callback
system
codes
interact
on
granular
level
represents
significant
advancement
toward
increased
modularity
among
scientific
codes.
second
Mechanic,
ecosystem
management
tool
utilizes
Docker
containerization
developing,
validating,
maintaining,
deploying
Additionally,
Mechanic
provides
framework
for
launching
simulations
in
each
interoperating
executed
separate
computational
environment.
This
eliminates
need
compile
multiple
production
single
environment,
reducing
opportunities
dependency
conflicts
lowering
barrier
entry
users
MDI-enabled
The Journal of Chemical Physics,
Год журнала:
2024,
Номер
161(2)
Опубликована: Июль 11, 2024
MiMiC
is
a
framework
for
performing
multiscale
simulations
in
which
loosely
coupled
external
programs
describe
individual
subsystems
at
different
resolutions
and
levels
of
theory.
To
make
it
highly
efficient
flexible,
we
adopt
an
interoperable
approach
based
on
multiple-program
multiple-data
(MPMD)
paradigm,
serving
as
intermediary
responsible
fast
data
exchange
interactions
between
the
subsystems.
The
main
goal
to
avoid
interfering
with
underlying
parallelization
programs,
including
operability
hybrid
architectures
(e.g.,
CPU/GPU),
keep
their
setup
execution
close
possible
original.
At
moment,
offers
implementation
electrostatic
embedding
quantum
mechanics/molecular
mechanics
(QM/MM)
that
has
demonstrated
unprecedented
parallel
scaling
large
biomolecules
using
CPMD
GROMACS
QM
MM
engines,
respectively.
However,
designed
high
flexibility
general
models
mind,
can
be
straightforwardly
extended
beyond
QM/MM.
In
this
article,
illustrate
software
design
features
framework,
compelling
choice
upcoming
era
exascale
high-performance
computing.
Journal of Chemical Theory and Computation,
Год журнала:
2024,
Номер
20(9), С. 3462 - 3472
Опубликована: Апрель 26, 2024
Adaptive
quantum
mechanics/molecular
mechanics
(QM/MM)
reclassifies
on-the-fly
a
molecule
or
molecular
fragment
as
QM
MM
during
dynamics
simulations
without
abrupt
changes
in
the
energy
forces.
Notably,
permuted
adaptive-partitioning
(PAP)
algorithms
have
been
applied
to
simulate
hydrated
proton,
with
mobile
zone
anchored
at
pseudoatom
called
proton
indicator.
The
position
of
indicator
approximates
location
delocalized
excess
yielding
smooth
trajectory
diffusing
via
Grotthuss
mechanism
aqueous
solutions.
zone,
which
has
taken
be
sphere
preset
radius,
follows
wherever
it
goes.
Although
are
successful,
use
spherical
one
disadvantage:
A
large
radius
must
utilized
minimize
chance
missing
water
molecules
that
important
translocation.
leads
is
computationally
expensive.
In
this
work,
we
report
new
way
set
up
where
includes
only
transfer.
importance
given
quantified
by
its
"weight"
depends
on
relation
reaction
path
weight
varies
smoothly,
ensuring
gradually
appears
disappears
from
changes,
required
PAP
method.
Consequently,
shape
evolves
on-the-fly,
keeping
small
possible
and
necessary.
Test
demonstrate
algorithm
significantly
improves
computation
efficiency
while
maintaining
proper
descriptions
transfer
bulk
water.