Peptide-Aware Chemical Language Model Successfully Predicts Membrane Diffusion of Cyclic Peptides
Journal of Chemical Information and Modeling,
Год журнала:
2025,
Номер
unknown
Опубликована: Янв. 8, 2025
Language
modeling
applied
to
biological
data
has
significantly
advanced
the
prediction
of
membrane
penetration
for
small-molecule
drugs
and
natural
peptides.
However,
accurately
predicting
diffusion
peptides
with
pharmacologically
relevant
modifications
remains
a
substantial
challenge.
Here,
we
introduce
PeptideCLM,
peptide-focused
chemical
language
model
capable
encoding
modifications,
unnatural
or
noncanonical
amino
acids,
cyclizations.
We
assess
this
by
cyclic
peptides,
demonstrating
greater
predictive
power
than
existing
models.
Our
is
versatile
can
be
extended
beyond
predictions
other
target
values.
Its
advantages
include
ability
macromolecules
using
string
notation,
largely
unexplored
domain,
simple,
flexible
architecture
that
allows
adaptation
any
peptide
macromolecule
set.
Язык: Английский
Prodrug Approach as a Strategy to Enhance Drug Permeability
Pharmaceuticals,
Год журнала:
2025,
Номер
18(3), С. 297 - 297
Опубликована: Фев. 21, 2025
Absorption
and
permeability
are
critical
physicochemical
parameters
that
must
be
balanced
to
achieve
optimal
drug
uptake.
These
key
factors
closely
linked
the
maximum
absorbable
dose
required
provide
appropriate
plasma
levels
of
drugs.
Among
various
strategies
employed
enhance
solubility
permeability,
prodrug
design
stands
out
as
a
highly
effective
versatile
approach
for
improving
properties
enabling
optimization
biopharmaceutical
pharmacokinetic
while
mitigating
adverse
effects.
Prodrugs
compounds
with
reduced
or
no
activity
that,
through
bio-reversible
chemical
enzymatic
processes,
release
an
active
parental
drug.
The
application
this
technology
has
led
significant
advancements
in
during
phase,
it
offers
broad
potential
further
development.
Notably,
approximately
13%
drugs
approved
by
U.S.
Food
Drug
Administration
(FDA)
between
2012
2022
were
prodrugs.
In
review
article,
we
will
explore
describing
examples
market
We
also
describe
use
optimize
PROteolysis
TArgeting
Chimeras
(PROTACs)
using
conjugation
technologies.
highlight
some
new
technologies
prodrugs
enrich
properties,
contributing
developing
safe
Язык: Английский
Cyclic peptide membrane permeability prediction using deep learning model based on molecular attention transformer
Frontiers in Bioinformatics,
Год журнала:
2025,
Номер
5
Опубликована: Март 11, 2025
Membrane
permeability
is
a
critical
bottleneck
in
the
development
of
cyclic
peptide
drugs.
Experimental
membrane
testing
costly,
and
precise
silico
prediction
tools
are
scarce.
In
this
study,
we
developed
CPMP
(
https://github.com/panda1103/CPMP
),
model
based
on
Molecular
Attention
Transformer
(MAT)
frame.
The
demonstrated
robust
predictive
performance,
achieving
determination
coefficients
R
2
)
0.67
for
PAMPA
prediction,
values
0.75,
0.62,
0.73
Caco-2,
RRCK,
MDCK
cell
predictions,
respectively.
Its
performance
outperforms
traditional
machine
learning
methods
graph-based
neural
network
models.
ablation
experiments,
validated
effectiveness
each
component
MAT
architecture.
Additionally,
analyzed
impact
data
pre-training
conformation
optimization
performance.
Язык: Английский
Peptide-specific chemical language model successfully predicts membrane diffusion of cyclic peptides
bioRxiv (Cold Spring Harbor Laboratory),
Год журнала:
2024,
Номер
unknown
Опубликована: Авг. 9, 2024
Language
modeling
applied
to
biological
data
has
significantly
advanced
the
prediction
of
membrane
penetration
for
small
molecule
drugs
and
natural
peptides.
However,
accurately
predicting
diffusion
peptides
with
pharmacologically
relevant
modifications
remains
a
substantial
challenge.
Here,
we
introduce
PeptideCLM,
peptide-focused
chemical
language
model
capable
encoding
modifications,
unnatural
or
non-canonical
amino
acids,
cyclizations.
We
assess
this
by
cyclic
peptides,
demonstrating
greater
predictive
power
than
existing
models.
Our
is
versatile
can
be
extended
beyond
predictions
other
target
values.
Its
advantages
include
ability
macromolecules
using
string
notation,
largely
unexplored
domain,
simple,
flexible
architecture
that
allows
adaptation
any
peptide
macromolecule
dataset.
Язык: Английский
Multi-Omic Approaches in Cancer-Related Micropeptide Identification
Proteomes,
Год журнала:
2024,
Номер
12(3), С. 26 - 26
Опубликована: Сен. 13, 2024
Despite
the
advances
in
modern
cancer
therapy,
malignant
diseases
are
still
a
leading
cause
of
morbidity
and
mortality
worldwide.
Conventional
treatment
methods
frequently
lead
to
side
effects
drug
resistance
patients,
highlighting
need
for
novel
therapeutic
approaches.
Recent
findings
have
identified
existence
non-canonical
micropeptides,
an
additional
layer
proteome
complexity,
also
called
microproteome.
These
small
peptides
promising
class
agents
with
potential
address
limitations
current
treatments.
The
microproteome
is
encoded
by
regions
genome
historically
annotated
as
non-coding,
its
has
been
revealed
thanks
recent
proteomic
bioinformatic
technology,
which
dramatically
improved
understanding
complexity.
Micropeptides
shown
be
biologically
active
several
types,
indicating
their
role.
Furthermore,
they
characterized
low
toxicity
high
target
specificity,
demonstrating
development
better
tolerated
drugs.
In
this
review,
we
survey
landscape
known
micropeptides
role
progression
or
treatment,
discuss
anticancer
agents,
describe
methodological
challenges
facing
field
research.
Язык: Английский