Most
hits
identified
in
the
drug
discovery
pipelines
and
even
40%
of
marketed
drugs
suffer
from
suboptimal
pharmacokinetic
profiles.
Co-
crystallization,
wherein
a
(or
candidate)
another
organic
molecule
form
multi-
component
crystal,
can
optimize
physicochemical
properties
those
molecules
without
hampering
their
pharmacological
activity.
However,
finding
promising
co-crystal
pairs
is
resource-intensive
due
to
vast
search
space.
Here
we
propose
DeepCrystal,
deep
learning
model
based
on
chemical
language
predict
co-crystallization.
We
rigorously
validate
DeepCrystal
find
that
it
achieves
78%
accuracy
realistic
settings
displays
superior
performance
existing
models.
Leveraging
represent
molecules,
estimate
uncertainty
its
predictions.
exploit
this
capability
challenging
prospective
study
discover
two
novel
diflunisal,
an
antiinflammatory
drug.
This
exemplifies
successful
application
accelerate
co-crystallization
process
lab,
highlighting
potential,
both
academic
industrial
settings.
International Journal of Molecular Sciences,
Journal Year:
2024,
Volume and Issue:
25(22), P. 12045 - 12045
Published: Nov. 9, 2024
Pharmaceutical
cocrystals
offer
a
versatile
approach
to
enhancing
the
properties
of
drug
compounds,
making
them
an
important
tool
in
formulation
and
development
by
improving
therapeutic
performance
patient
experience
pharmaceutical
products.
The
prediction
involves
using
computational
theoretical
methods
identify
potential
cocrystal
formers
understand
interactions
between
active
ingredient
coformers.
This
process
aims
predict
whether
two
or
more
molecules
can
form
stable
structure
before
performing
experimental
synthesis,
thus
saving
time
resources.
In
this
review,
commonly
used
are
first
overviewed
then
evaluated
based
on
three
criteria:
efficiency,
cost-effectiveness,
user-friendliness.
Based
these
considerations,
we
suggest
researchers
without
strong
experiences
which
tools
should
be
tested
as
step
workflow
rational
design
cocrystals.
However,
optimal
choice
depends
specific
needs
resources,
combining
from
different
categories
powerful
approach.
Burger's Medicinal Chemistry and Drug Discovery,
Journal Year:
2025,
Volume and Issue:
unknown, P. 1 - 57
Published: May 26, 2025
Abstract
Using
artificial
intelligence
and
machine
learning,
computational
tools
are
increasingly
accelerating
new
medicine
discovery.
Applications
to
the
stages
of
drug
discovery,
including
target
identification,
hit
lead
optimization,
developability
assessment,
described.
This
chapter
provides
a
compilation
databases,
software
packages,
web‐based
applications
for
evaluation,
optimization
molecules.
Pharmaceutics,
Journal Year:
2023,
Volume and Issue:
16(1), P. 27 - 27
Published: Dec. 24, 2023
This
review
discusses
the
entire
progress
made
on
anthelmintic
drug
praziquantel,
focusing
solid
state
and,
therefore,
anhydrous
crystalline
polymorphs,
amorphous
forms,
and
multicomponent
systems
(i.e.,
hydrates,
solvates,
cocrystals).
Despite
having
been
extensively
studied
over
last
50
years,
new
polymorphs
greater
part
of
their
cocrystals
have
only
identified
in
past
decade.
Progress
crystal
engineering
science
(e.g.,
use
mechanochemistry
as
a
form
screening
tool
more
strategic
structure-based
methods),
along
with
development
analytical
techniques,
including
Synchrotron
X-ray
analyses,
spectroscopy,
microscopy,
furthered
identification
unknown
structures
drug.
Also,
computational
modeling
has
significantly
contributed
to
prediction
design
by
considering
structural
conformations
interactions
energy.
Whilst
insights
praziquantel
discussed
present
will
give
significant
contribution
controlling
formation
during
manufacturing
formulation,
detailed
forms
help
designing
implementing
future
praziquantel-based
functional
materials.
The
latter
hopefully
overcome
praziquantel’s
numerous
drawbacks
exploit
its
potential
field
neglected
tropical
diseases.
Journal of Computational Chemistry,
Journal Year:
2024,
Volume and Issue:
45(29), P. 2465 - 2475
Published: July 3, 2024
Abstract
Cocrystals
are
assemblies
of
more
than
one
type
molecule
stabilized
through
noncovalent
interactions.
They
promising
materials
for
improved
drug
formulation
in
which
the
stability,
solubility,
or
biocompatibility
active
pharmaceutical
ingredient
(API)
is
by
including
a
coformer.
In
this
work,
range
density
functional
theory
(DFT)
and
tight
binding
(DFTB)
models
systematically
compared
their
ability
to
predict
lattice
enthalpy
broad
existing
pharmaceutically
relevant
cocrystals.
These
from
cocrystals
containing
model
compounds
4,4′‐bipyridine
oxalic
acid
those
with
well
benchmarked
APIs
aspirin
paracetamol,
all
tested
large
set
alternative
coformers.
For
simple
cocrystals,
there
general
consensus
calculated
different
DFT
models.
API
coformers
predictions
depend
strongly
on
model.
The
significantly
lighter
DFTB
unrealistic
values
even
Approximately
40%
of
marketed
drugs
exhibit
suboptimal
pharmacokinetic
profiles.
Co-crystallization,
where
pairs
molecules
form
a
multicomponent
crystal,
constitutes
promising
strategy
to
enhance
physicochemical
properties
without
compromising
the
pharmacological
activity.
However,
finding
co-crystal
is
resource-intensive,
due
vast
number
possible
combinations.
We
present
DeepCocrystal,
novel
deep
learning
approach
designed
predict
formation
by
processing
'chemical
language'
from
supramolecular
vantage
point.
Rigorous
validation
DeepCocrystal
showed
balanced
accuracy
78%
in
realistic
scenarios,
outperforming
existing
models.
By
leveraging
molecular
string
representations,
can
also
estimate
uncertainty
its
predictions.
harness
this
capability
challenging
prospective
study,
and
successfully
discovered
two
diflunisal,
an
anti-inflammatory
drug.
This
study
underscores
potential
--
particular
chemical
language
accelerate
co-crystallization,
ultimately
drug
development,
both
academic
industrial
contexts.