Journal of Chemical Information and Modeling,
Год журнала:
2024,
Номер
64(11), С. 4419 - 4425
Опубликована: Май 17, 2024
The
atomic
partial
charge
is
of
great
importance
in
many
fields,
such
as
chemistry
and
drug-target
recognition.
However,
conventional
quantum-based
computing
charges
relatively
slow,
limiting
further
applications
analysis.
With
the
help
machine
learning
methods,
various
kinds
models
appear
to
speed
up
calculations.
there
are
still
some
concerning
problems.
Some
based
on
geometric
coordinates
require
high-accuracy
geometry
optimization
a
preprocess,
while
other
have
limitation
size
input
molecules
that
narrow
model.
Here,
we
propose
model
message-passing
featurizer.
This
preprocessing
featurizer
can
quickly
extract
environment
information
from
molecule
according
connectivity
inside
molecule.
resulting
descriptor
be
used
with
neural
network
predict
charge.
able
automatically
adapt
any
remaining
efficient
achieves
root-mean-square
error
Hirshfeld
prediction
0.018e,
an
overall
time
complexity
O(n2).
Thus,
this
could
enlarge
range
more
fields
cases.
Wiley Interdisciplinary Reviews Computational Molecular Science,
Год журнала:
2025,
Номер
15(2)
Опубликована: Март 1, 2025
ABSTRACT
Proteolysis
targeting
chimera
(PROTAC)
induces
specific
protein
degradation
through
the
ubiquitin–proteasome
system
and
offers
significant
advantages
over
small
molecule
drugs.
They
are
emerging
as
a
promising
avenue,
particularly
in
previously
“undruggable”
targets.
Traditional
PROTACs
have
been
discovered
large‐scale
experimental
screening.
Extensive
research
efforts
focused
on
unraveling
biological
pharmacological
functions
of
PROTACs,
with
strides
made
toward
transitioning
from
empirical
discovery
to
rational,
structure‐based
design
strategies.
This
review
provides
an
overview
recent
representative
computer‐aided
drug
studies
PROTACs.
We
highlight
how
utilization
targeted
database,
molecular
modeling
techniques,
machine
learning
algorithms,
computational
methods
contributes
facilitating
PROTAC
discovery.
Furthermore,
we
conclude
achievements
field
explore
challenges
future
directions.
aim
offer
insights
references
for
rational
BMC Medical Genomics,
Год журнала:
2025,
Номер
18(1)
Опубликована: Апрель 9, 2025
Coronary
Artery
Disease
(CAD)
is
the
most
common
cardiovascular
disease
worldwide,
threatening
human
health,
quality
of
life
and
longevity.
Aging
a
dominant
risk
factor
for
CAD.
This
study
aims
to
investigate
potential
mechanisms
aging-related
genes
CAD,
make
molecular
drug
predictions
that
will
contribute
diagnosis
treatment.
We
downloaded
gene
expression
profile
circulating
leukocytes
in
CAD
patients
(GSE12288)
from
Gene
Expression
Omnibus
database,
obtained
differentially
expressed
aging
through
"limma"
package
GenaCards
tested
their
biological
functions.
Further
screening
related
characteristic
(ARCGs)
using
least
absolute
shrinkage
selection
operator
random
forest,
generating
nomogram
charts
ROC
curves
evaluating
diagnostic
efficacy.
Immune
cells
were
estimated
by
ssGSEA,
then
combine
ARCGs
with
immune
clinical
indicators
based
on
Pearson
correlation
analysis.
Unsupervised
cluster
analysis
was
used
construct
clusters
assess
functional
characteristics
between
clusters.
The
DSigDB
database
employed
explore
targeted
drugs
ARCGs,
docking
carried
out
Autodock
Vina.
Finally,
single-cell
data
(GSE159677)
arterial
intima
further
signature
different
cell
subpopulations.
identified
8
associated
which
HIF1A
FGFR3
up
while
NOX4,
TCF7L2,
HK3,
CDK18,
TFAP4,
ITPK1
down
patients.
Based
this,
can
be
divided
into
two
clusters,
among
A
mainly
involves
pathways
such
as
ECM
receptor
interaction
focal
adhesion;
B
amimo
sugar
nucleotide
metabolism
pyrimidine
metabolism.
In
addition,
results
showed
retinoic
acid
resveratrol
had
good
binding
affinity
targets
genes.
ITPK1,
specifically
types
atherosclerotic
tissues.
Our
several
may
involved
pathogenesis
progression
Further,
candidate
molecule
inhibiting
these
targets.
Journal of the American Chemical Society,
Год журнала:
2025,
Номер
unknown
Опубликована: Апрель 11, 2025
Understanding
protein-drug
complex
structures
is
crucial
for
elucidating
therapeutic
mechanisms
and
side
effects.
Blind
docking
facilitates
site
identification
but
hindered
by
computational
complexity
imprecise
scoring,
causing
ambiguity.
Dipolar
electron
paramagnetic
resonance
(EPR)
provides
spin-spin
distances
struggles
to
determine
relative
positions
within
complexes.
We
present
a
novel
approach
combining
GPU-accelerated
blind
with
EPR
distance
constraints
enhance
binding
detection.
Our
algorithm
uses
single
distribution
filter
validate
results.
Ligand
poses
from
are
clustered,
filtered
expected
distances,
refined
through
focused
docking.
To
illustrate
our
approach,
we
investigated
human
serum
albumin
porphyrin-based
photosensitizers
used
in
photodynamic
therapy.
Combining
EPR,
identified
possible
sites,
demonstrating
that
data
significantly
reduce
configurations
provide
experimentally
validated
information.
This
strategy
produces
detailed
map
of
photoligand
revealing
may
occur
away
standard
sites
often
involves
multiple
locations.
Furthermore,
it
overcomes
key
limitations
fluorescence-based
methods,
which
prone
misinterpretation
studies
due
non
one-to-one
donor-acceptor
relationships.
By
resolving
ambiguities
both
framework
versatile
platform
investigating
EPR-active
ligands.
bioRxiv (Cold Spring Harbor Laboratory),
Год журнала:
2025,
Номер
unknown
Опубликована: Апрель 18, 2025
Abstract
Developing
a
physical
understanding
of
the
interactions
between
macro-molecular
target
and
its
ligands
is
crucial
step
in
structure-based
drug
design.
Although
many
tools
exist
to
characterize
protein-binding
pockets
silico,
this
not
yet
case
for
RNA,
which
has
only
recently
been
recognized
as
suitable
small
ligands.
Molecular
Interaction
Fields
(MIF)
are
useful
tool
given
binding
pocket.
However,
classical
MIFs
heavily
rely
on
use
probes,
makes
their
calculations
accurate
but
very
specific
partners
question.
We
develop
here
simple
version
MIF,
that
we
call
Statistical
(SMIF),
based
functional
forms
inspired
by
coarse-grained
models
parametrized
PDB
structures
previous
statistical
analysis
main
form
typical
macromolecules,
namely
hydrogen
bonding,
stacking,
hydrophobic
interactions.
show
these
fields,
despite
simplicity,
informative
overall
agreement
with
pharmacophoric
models.
Thanks
carefully
optimized
code,
our
fast
can
be
performed
bulk
large
set
or
even
full
macromolecule.
As
shown
few
representative
examples,
latter
possibility
opens
way
systems
20
80
k
atoms
relation
surrounding
environment,
i.e.,
lipidic
membrane,
ligand,
another
macromolecular
partner,
allowing
detailed
visualization
possible
Intrinsically
disordered
proteins
(IDPs)
and
biomolecular
condensates
are
critical
for
cellular
processes
physiological
functions.
Abnormal
can
cause
diseases
such
as
cancer
neurodegenerative
disorders.
IDPs,
including
intrinsically
regions
(IDRs),
were
previously
considered
undruggable
due
to
their
lack
of
stable
binding
pockets.
However,
recent
evidence
indicates
that
targeting
them
influence
processes.
This
review
explores
current
strategies
target
IDPs
condensates,
potential
improvements,
the
challenges
opportunities
in
this
evolving
field.
Research Square (Research Square),
Год журнала:
2025,
Номер
unknown
Опубликована: Май 7, 2025
Abstract
The
aim
of
this
research
is
to
determine
the
phytochemicals
and
characterize
cannabis
sativa
leaf
extracts
because
they
are
one
consumed
with
HPLC,
GC-MS,
VV-VIS,
FT-IR
AAS.
phytochemical
screening
showed
that
phenols
tannins
ware
determinant
phytochemicals,
analyzed
functional
group
present,
HPLC
Catechin
acid
di-hydrate
was
major
compound
in
dichloromethane
extract
ethanolic
extract,
while
Naringin
component
aqueous
GC-MS
palmitic
,
Linoleic
7-oxtadecanoic
Linolenic
it
confirmed
by
UV-VIS
AAS
Cannabis
sativa
L.
found
contain
higher
levels
heavy
metals
Fe,
Zn,
Mn,
Cu.
