Nucleic Acids Research,
Год журнала:
2024,
Номер
53(D1), С. D1633 - D1644
Опубликована: Ноя. 22, 2024
Abstract
BindingDB
(bindingdb.org)
is
a
public,
web-accessible
database
of
experimentally
measured
binding
affinities
between
small
molecules
and
proteins,
which
supports
diverse
applications
including
medicinal
chemistry,
biochemical
pathway
annotation,
training
artificial
intelligence
models
computational
chemistry
methods
development.
This
update
reports
significant
growth
enhancements
since
our
last
review
in
2016.
Of
note,
the
now
contains
2.9
million
measurements
spanning
1.3
compounds
thousands
protein
targets.
largely
attributable
to
unique
focus
on
curating
data
from
US
patents,
has
yielded
substantial
influx
novel
data.
Recent
improvements
include
remake
website
following
responsive
web
design
principles,
enhanced
search
filtering
capabilities,
new
download
options
webservices
establishment
long-term
archive
replicated
across
dispersed
sites.
We
also
discuss
BindingDB’s
positioning
relative
related
resources,
its
open
sharing
policies,
insights
gleaned
dataset
plans
for
future
Chemical Science,
Год журнала:
2024,
Номер
15(11), С. 4146 - 4160
Опубликована: Янв. 1, 2024
Reinforcement
learning
(RL)
is
a
powerful
and
flexible
paradigm
for
searching
solutions
in
high-dimensional
action
spaces.
However,
bridging
the
gap
between
playing
computer
games
with
thousands
of
simulated
episodes
solving
real
scientific
problems
complex
involved
environments
(up
to
actual
laboratory
experiments)
requires
improvements
terms
sample
efficiency
make
most
expensive
information.
The
discovery
new
drugs
major
commercial
application
RL,
motivated
by
very
large
nature
chemical
space
need
perform
multiparameter
optimization
(MPO)
across
different
properties.
Journal of the American Chemical Society,
Год журнала:
2024,
Номер
146(29), С. 19654 - 19659
Опубликована: Июль 11, 2024
We
evaluate
the
effectiveness
of
pretrained
and
fine-tuned
large
language
models
(LLMs)
for
predicting
synthesizability
inorganic
compounds
selection
precursors
needed
to
perform
synthesis.
The
predictions
LLMs
are
comparable
to─and
sometimes
better
than─recent
bespoke
machine
learning
these
tasks
but
require
only
minimal
user
expertise,
cost,
time
develop.
Therefore,
this
strategy
can
serve
both
as
an
effective
strong
baseline
future
studies
various
chemical
applications
a
practical
tool
experimental
chemists.
Nature Communications,
Год журнала:
2024,
Номер
15(1)
Опубликована: Июль 25, 2024
The
prefusion
conformation
of
human
metapneumovirus
fusion
protein
(hMPV
Pre-F)
is
critical
for
eliciting
the
most
potent
neutralizing
antibodies
and
preferred
immunogen
an
efficacious
vaccine
against
hMPV
respiratory
infections.
Here
we
show
that
additional
cleavage
event
in
F
allows
closure
correct
folding
trimer.
We
therefore
engineered
to
undergo
double
cleavage,
which
enabled
screening
Pre-F
stabilizing
substitutions
at
natively
folded
protomer
interfaces.
To
identify
these
substitutions,
developed
AI
convolutional
classifier
successfully
predicts
complex
polar
interactions
often
overlooked
by
physics-based
methods
visual
inspection.
combination
processing,
stabilization
interface
regions
membrane-proximal
stem,
resulted
a
candidate
without
need
heterologous
trimerization
domain
exhibited
high
expression
yields
thermostability.
Cryo-EM
analysis
shows
complete
ectodomain
structure,
including
specific
interaction
newly
identified
cleaved
C-terminus
with
adjacent
protomer.
Importantly,
induces
cross-neutralizing
antibody
responses
resulting
near
protection
challenge
cotton
rats,
making
highly
stable,
double-cleaved
trimer
attractive
candidate.
Nucleic Acids Research,
Год журнала:
2024,
Номер
53(D1), С. D1633 - D1644
Опубликована: Ноя. 22, 2024
Abstract
BindingDB
(bindingdb.org)
is
a
public,
web-accessible
database
of
experimentally
measured
binding
affinities
between
small
molecules
and
proteins,
which
supports
diverse
applications
including
medicinal
chemistry,
biochemical
pathway
annotation,
training
artificial
intelligence
models
computational
chemistry
methods
development.
This
update
reports
significant
growth
enhancements
since
our
last
review
in
2016.
Of
note,
the
now
contains
2.9
million
measurements
spanning
1.3
compounds
thousands
protein
targets.
largely
attributable
to
unique
focus
on
curating
data
from
US
patents,
has
yielded
substantial
influx
novel
data.
Recent
improvements
include
remake
website
following
responsive
web
design
principles,
enhanced
search
filtering
capabilities,
new
download
options
webservices
establishment
long-term
archive
replicated
across
dispersed
sites.
We
also
discuss
BindingDB’s
positioning
relative
related
resources,
its
open
sharing
policies,
insights
gleaned
dataset
plans
for
future
