Journal of Chemical Information and Modeling,
Год журнала:
2025,
Номер
unknown
Опубликована: Фев. 10, 2025
Antimicrobial
peptides
are
a
promising
strategy
to
combat
antimicrobial
resistance.
However,
the
experimental
discovery
of
is
both
time-consuming
and
laborious.
In
recent
years,
development
computational
technologies
(especially
deep
learning)
has
provided
new
opportunities
for
peptide
prediction.
Various
models
have
been
proposed
predict
peptide.
this
review,
we
focus
on
learning
We
first
collected
summarized
available
data
resources
peptides.
Subsequently,
existing
discussed
their
limitations
challenges.
This
study
aims
help
biologists
design
better
A
promoter
is
a
specific
sequence
in
DNA
that
has
transcriptional
regulatory
functions,
playing
role
initiating
gene
expression.
Identifying
promoters
and
their
strengths
can
provide
valuable
information
related
to
human
diseases.
In
recent
years,
computational
methods
have
gained
prominence
as
an
effective
means
for
identifying
promoter,
offering
more
efficient
alternative
labor-intensive
biological
approaches.
Abstract
Motivation
Peptides
are
promising
agents
for
the
treatment
of
a
variety
diseases
due
to
their
specificity
and
efficacy.
However,
development
peptide-based
drugs
is
often
hindered
by
potential
toxicity
peptides,
which
poses
significant
barrier
clinical
application.
Traditional
experimental
methods
evaluating
peptide
time-consuming
costly,
making
process
inefficient.
Therefore,
there
an
urgent
need
computational
tools
specifically
designed
predict
accurately
rapidly,
facilitating
identification
safe
candidates
drug
development.
Results
We
provide
here
novel
approach,
CAPTP,
leverages
power
convolutional
self-attention
enhance
prediction
from
amino
acid
sequences.
CAPTP
demonstrates
outstanding
performance,
achieving
Matthews
correlation
coefficient
approximately
0.82
in
both
cross-validation
settings
on
independent
test
datasets.
This
performance
surpasses
that
existing
state-of-the-art
predictors.
Importantly,
maintains
its
robustness
generalizability
even
when
dealing
with
data
imbalances.
Further
analysis
reveals
certain
sequential
patterns,
particularly
head
central
regions
crucial
determining
toxicity.
insight
can
significantly
inform
guide
design
safer
drugs.
Availability
implementation
The
source
code
freely
available
at
https://github.com/jiaoshihu/CAPTP.
In
biological
organisms,
metal
ion-binding
proteins
participate
in
numerous
metabolic
activities
and
are
closely
associated
with
various
diseases.
To
accurately
predict
whether
a
protein
binds
to
ions
the
type
of
protein,
this
study
proposed
classifier
named
MIBPred.
The
incorporated
advanced
Word2Vec
technology
from
field
natural
language
processing
extract
semantic
features
sequence
combined
them
position-specific
score
matrix
(PSSM)
features.
Furthermore,
an
ensemble
learning
model
was
employed
for
classification
task.
model,
we
independently
trained
XGBoost,
LightGBM,
CatBoost
algorithms
integrated
output
results
through
SVM
voting
mechanism.
This
innovative
combination
has
led
significant
breakthrough
predictive
performance
our
model.
As
result,
achieved
accuracies
95.13%
85.19%,
respectively,
predicting
their
types.
Our
research
not
only
confirms
effectiveness
extracting
information
sequences
but
also
highlights
outstanding
MIBPred
problem
provides
reliable
tool
method
in-depth
exploration
structure
function
proteins.
Briefings in Bioinformatics,
Год журнала:
2023,
Номер
25(1)
Опубликована: Ноя. 22, 2023
Abstract
As
a
kind
of
small
molecule
protein
that
can
fight
against
various
microorganisms
in
nature,
antimicrobial
peptides
(AMPs)
play
an
indispensable
role
maintaining
the
health
organisms
and
fortifying
defenses
diseases.
Nevertheless,
experimental
approaches
for
AMP
identification
still
demand
substantial
allocation
human
resources
material
inputs.
Alternatively,
computing
assist
researchers
effectively
promptly
predict
AMPs.
In
this
study,
we
present
novel
predictor
called
iAMP-Attenpred.
far
as
know,
is
first
work
not
only
employs
popular
BERT
model
field
natural
language
processing
(NLP)
AMPs
feature
encoding,
but
also
utilizes
idea
combining
multiple
models
to
discover
Firstly,
treat
each
amino
acid
from
preprocessed
non-AMP
sequences
word,
then
input
it
into
pre-training
extraction.
Moreover,
features
obtained
method
are
fed
composite
composed
one-dimensional
CNN,
BiLSTM
attention
mechanism
better
discriminating
features.
Finally,
flatten
layer
fully
connected
layers
utilized
final
classification
Experimental
results
reveal
that,
compared
with
existing
predictors,
our
iAMP-Attenpred
achieves
performance
indicators,
such
accuracy,
precision
so
on.
This
further
demonstrates
using
approach
capture
effective
information
peptide
deep
learning
meaningful
predicting
