Research Square (Research Square),
Год журнала:
2024,
Номер
unknown
Опубликована: Фев. 9, 2024
Abstract
With
the
rapid
advancements
in
molecular
biology
and
genomics,
a
multitude
of
connections
between
RNA
diseases
has
been
unveiled,
making
efficient
accurate
extraction
RNA-disease
relationships
(RD
relationships)
from
extensive
biomedical
literature
crucial
for
advancing
research
this
field.
This
study
introduces
RDscan,
novel
text
mining
method
developed
based
on
pre-training
fine-tuning
strategy,
aimed
at
automatically
extracting
RD-related
information
vast
corpus
using
pre-trained
large
language
models.
Initially,
we
constructed
dedicated
RD
corpus,
comprising
2,082
positive
2,000
negative
statements,
alongside
an
independent
test
dataset
training
evaluating
RDscan.
Subsequently,
by
Bioformer
BioBERT
models,
RDscan
demonstrated
exceptional
performance
classification
named
entity
recognition
(NER)
tasks.
In
5-fold
cross-validation,
significantly
outperformed
traditional
machine
learning
methods.
summary,
represents
first
tool
specifically
designed
relationship
extraction,
is
freely
available
https://github.com/ZhangCellab/RDScaning.
Diagnostics,
Год журнала:
2023,
Номер
13(14), С. 2465 - 2465
Опубликована: Июль 24, 2023
Heparin-binding
protein
(HBP)
is
a
cationic
antibacterial
derived
from
multinuclear
neutrophils
and
an
important
biomarker
of
infectious
diseases.
The
correct
identification
HBP
great
significance
to
the
study
This
work
provides
first
recognition
framework
based
on
machine
learning
accurately
identify
HBP.
By
using
four
sequence
descriptors,
non-HBP
samples
were
represented
by
discrete
numbers.
inputting
these
features
into
support
vector
(SVM)
random
forest
(RF)
algorithm
comparing
prediction
performances
methods
training
data
independent
test
data,
it
found
that
SVM-based
classifier
has
greatest
potential
model
could
produce
auROC
0.981
±
0.028
10-fold
cross-validation
overall
accuracy
95.0%
data.
As
for
recognition,
will
provide
some
help
diseases
stimulate
further
research
in
related
fields.
Gastric
cancer
(GC)
is
a
prominent
contributor
to
global
cancer‑related
mortalities,
and
deeper
understanding
of
its
molecular
characteristics
tumor
heterogeneity
required.
Single‑cell
omics
spatial
transcriptomics
(ST)
technologies
have
revolutionized
research
by
enabling
the
exploration
cellular
landscapes
at
single‑cell
level.
In
present
review,
an
overview
advancements
in
ST
their
applications
GC
provided.
Firstly,
multiple
methods
are
discussed,
highlighting
ability
offer
unique
insights
into
gene
expression,
genetic
alterations,
epigenomic
modifications,
protein
expression
patterns
location
tissues.
Furthermore,
summary
provided
key
findings
from
previous
on
used
GC,
which
valuable
diagnosis
prognosis,
microenvironment
analysis,
treatment
response.
summary,
application
has
revealed
levels
holds
promise
for
improving
diagnostics,
personalized
treatments
patient
outcomes
GC.
In
China,
traditional
Chinese
medicine
(TCM)
has
been
widely
used
for
coronavirus
infectious
disease
2019
(COVID-19)
prevention,
treatment,
and
recovery
played
a
part
in
the
battle
against
disease.
A
variety
of
TCM
treatments
have
recommended
different
stages
COVID-19.
But,
to
best
our
knowledge,
comprehensive
database
storing
organizing
anti-COVID
is
still
lacking.
Herein,
we
developed
TCM2COVID,
manually
curated
resource
formulas,
natural
products
(NPs),
herbs.
The
current
version
TCM2COVID
(1)
documents
over
280
formulas
(including
300
herbs)
with
detailed
clinical
evidence
therapeutic
mechanism
information;
(2)
records
80
NPs
potential
mechanisms;
(3)
launches
useful
web
server
querying,
analyzing
visualizing
documented
similar
those
supplied
by
user
(formula
similarity
analysis).
summary,
TCM2COVD
provides
user-friendly
practical
platform
documenting,
browsing
treatments,
will
help
development
elucidation
mechanisms
action
new
therapies
support
fight
COVID-19
epidemic.
freely
available
at
http://zhangy-lab.cn/tcm2covid/.
Journal of Cellular and Molecular Medicine,
Год журнала:
2024,
Номер
28(13)
Опубликована: Июль 1, 2024
Both
osteoporosis
and
tendinopathy
are
widely
prevalent
disorders,
encountered
in
diverse
medical
contexts.
Whilst
each
condition
has
distinct
pathophysiological
characteristics,
they
share
several
risk
factors
underlying
causes.
Notably,
oxidative
stress
emerges
as
a
crucial
intersecting
factor,
playing
pivotal
role
the
onset
progression
of
both
diseases.
This
imbalance
arises
from
dysregulation
generating
neutralising
reactive
oxygen
species
(ROS),
leading
to
an
abnormal
environment.
Elevated
levels
ROS
can
induce
multiple
cellular
disruptions,
such
cytotoxicity,
apoptosis
activation
reduced
cell
function,
contributing
tissue
deterioration
weakening
structural
integrity
bones
tendons.
Antioxidants
substances
that
prevent
or
slow
down
oxidation
process,
including
Vitamin
C,
melatonin,
resveratrol,
anthocyanins
so
on,
demonstrating
potential
treating
these
overlapping
disorders.
comprehensive
review
aims
elucidate
complex
within
interlinked
pathways
comorbid
conditions.
By
integrating
contemporary
research
empirical
findings,
our
objective
is
outline
new
conceptual
models
innovative
treatment
strategies
for
effectively
managing
underscores
importance
further
in-depth
validate
efficacy
antioxidants
traditional
Chinese
medicine
plans,
well
explore
targeted
interventions
focused
on
promising
areas
future
advancements.
BMC Complementary Medicine and Therapies,
Год журнала:
2025,
Номер
25(1)
Опубликована: Янв. 8, 2025
Abstract
Background
The
integration
of
traditional
Chinese
medicine
(TCM)
into
emergency
health
systems
in
China
serves
as
a
model
for
global
policy
development
and
refining
the
inclusion
emergencies.
Methods
This
study
investigated
13
public
policies
related
to
TCM
released
by
central
government
from
2003–2023.
A
PMC(Policy
Modeling
Consistency)
index
was
developed
combining
ROSTCM
text
mining
analysis
software.
contents
these
documents
were
quantitatively
assessed
using
10
first-
40
s-level
indicators.
Results
content
results
showed
that
current
focus
on
treatment,
State
Administration
Traditional
Medicine
is
issuing
authority
main
policies,
most
which
are
issued
form
notice.
scoring
two,
five,
three,
three
rated
excellent,
good,
qualified,
unqualified,
respectively.
indicates
quality
use
response
normally
distributed
generally
although
room
further
improvement
exists;
should
follow
principles
science,
reasonableness,
operability,
be
updated
timely
manner
with
continuous
governance
period
while
focusing
content,
safeguards,
role
measures.
Conclusion
Effective
backed
state
institutions
vital.
includes
enforcing
relevant
laws
regulations,
establishing
multidisciplinary
medical
teams,
developing
integrated
strategies
support
clinical
research
maximize
unique
benefits
medicine.
Information,
Год журнала:
2024,
Номер
15(3), С. 163 - 163
Опубликована: Март 13, 2024
Accurate
prediction
of
subcellular
localization
viral
proteins
is
crucial
for
understanding
their
functions
and
developing
effective
antiviral
drugs.
However,
this
task
poses
a
significant
challenge,
especially
when
relying
on
expensive
time-consuming
classical
biological
experiments.
In
study,
we
introduced
computational
model
called
E-MuLA,
based
deep
learning
network
that
combines
multiple
local
attention
modules
to
enhance
feature
extraction
from
protein
sequences.
The
superior
performance
the
E-MuLA
has
been
demonstrated
through
extensive
comparisons
with
LSTM,
CNN,
AdaBoost,
decision
trees,
KNN,
other
state-of-the-art
methods.
It
noteworthy
achieved
an
accuracy
94.87%,
specificity
98.81%,
sensitivity
84.18%,
indicating
potential
become
tool
predicting
virus
localization.