Molecular Cell,
Journal Year:
2021,
Volume and Issue:
81(10), P. 2216 - 2230.e10
Published: April 13, 2021
DNA
double-strand
break
(DSB)
repair
is
mediated
by
multiple
pathways.
It
thought
that
the
local
chromatin
context
affects
pathway
choice,
but
underlying
principles
are
poorly
understood.
Using
a
multiplexed
reporter
assay
in
combination
with
Cas9
cutting,
we
systematically
measure
relative
activities
of
three
DSB
pathways
as
function
>1,000
genomic
locations.
This
reveals
non-homologous
end-joining
(NHEJ)
broadly
biased
toward
euchromatin,
while
contribution
microhomology-mediated
(MMEJ)
higher
specific
heterochromatin
contexts.
In
H3K27me3-marked
heterochromatin,
inhibition
H3K27
methyltransferase
EZH2
reverts
balance
NHEJ.
Single-stranded
template
(SSTR),
often
used
for
precise
CRISPR
editing,
competes
MMEJ
and
moderately
linked
to
context.
These
results
provide
insight
into
impact
on
guidance
design
Cas9-mediated
genome
editing
experiments.
Hundreds
of
human
genes
are
associated
with
neurological
diseases,
but
translation
into
tractable
biological
mechanisms
is
lagging.
Larval
zebrafish
an
attractive
model
to
investigate
genetic
contributions
diseases.
However,
current
CRISPR-Cas9
methods
difficult
apply
large
screens
studying
behavioural
phenotypes.
To
facilitate
rapid
screening,
we
developed
a
simple
sequencing-free
tool
validate
gRNAs
and
highly
effective
method
capable
converting
>90%
injected
embryos
directly
F0
biallelic
knockouts.
We
demonstrate
that
knockouts
reliably
recapitulate
complex
mutant
phenotypes,
such
as
altered
molecular
rhythms
the
circadian
clock,
escape
responses
irritants,
multi-parameter
day-night
locomotor
behaviours.
The
technique
sufficiently
robust
knockout
multiple
in
same
animal,
for
example
create
transparent
triple
crystal
fish
imaging.
Our
cuts
experimental
time
from
gene
phenotype
months
one
week.
Nature Communications,
Journal Year:
2022,
Volume and Issue:
13(1)
Published: April 1, 2022
Deep
Learning
(DL)
has
recently
enabled
unprecedented
advances
in
one
of
the
grand
challenges
computational
biology:
half-century-old
problem
protein
structure
prediction.
In
this
paper
we
discuss
recent
advances,
limitations,
and
future
perspectives
DL
on
five
broad
areas:
prediction,
function
genome
engineering,
systems
biology
data
integration,
phylogenetic
inference.
We
each
application
area
cover
main
bottlenecks
approaches,
such
as
training
data,
scope,
ability
to
leverage
existing
architectures
new
contexts.
To
conclude,
provide
a
summary
subject-specific
general
for
across
biosciences.
Molecular Cell,
Journal Year:
2018,
Volume and Issue:
73(4), P. 699 - 713.e6
Published: Dec. 13, 2018
The
CRISPR-Cas9
system
has
successfully
been
adapted
to
edit
the
genome
of
various
organisms.
However,
our
ability
predict
editing
outcome
at
specific
sites
is
limited.
Here,
we
examined
indel
profiles
over
1,000
genomic
in
human
cells
and
uncovered
general
principles
guiding
CRISPR-mediated
DNA
editing.
We
find
that
precision
(i.e.,
recurrence
a
indel)
varies
considerably
among
sites,
with
some
targets
showing
one
highly
preferred
others
displaying
numerous
infrequent
indels.
Editing
correlates
efficiency
preference
for
single-nucleotide
homologous
insertions.
Precise
can
be
predicted
based
on
simple
rules
mainly
depend
fourth
nucleotide
upstream
protospacer
adjacent
motif
(PAM).
Indel
are
robust,
but
they
influenced
by
chromatin
features.
Our
findings
have
important
implications
clinical
applications
CRISPR
technology
reveal
patterns
broken
end
joining
provide
insights
into
repair
mechanisms.
Nucleic Acids Research,
Journal Year:
2019,
Volume and Issue:
47(15), P. 7989 - 8003
Published: May 24, 2019
Non-homologous
end-joining
(NHEJ)
plays
an
important
role
in
double-strand
break
(DSB)
repair
of
DNA.
Recent
studies
have
shown
that
the
error
patterns
NHEJ
are
strongly
biased
by
sequence
context,
but
these
were
based
on
relatively
few
templates.
To
investigate
this
more
thoroughly,
we
systematically
profiled
∼1.16
million
independent
mutational
events
resulting
from
CRISPR/Cas9-mediated
cleavage
and
NHEJ-mediated
DSB
6872
synthetic
target
sequences,
introduced
into
a
human
cell
line
via
lentiviral
infection.
We
find
that:
(i)
insertions
dominated
1
bp
templated
immediately
upstream
site,
(ii)
deletions
predominantly
associated
with
microhomology
(iii)
targets
exhibit
variable
reproducible
diversity
respect
to
number
relative
frequency
outcomes
which
they
give
rise.
From
data,
trained
model
uses
local
context
predict
distribution
outcomes.
Exploiting
bias
towards
mediated
events,
demonstrate
programming
deletion
introducing
specific
locations
vicinity
site.
anticipate
our
results
will
inform
investigations
mechanisms
as
well
design
CRISPR/Cas9
experiments
for
diverse
applications
including
genome-wide
screens,
gene
therapy,
lineage
tracing
molecular
recording.
Science Advances,
Journal Year:
2019,
Volume and Issue:
5(11)
Published: Nov. 1, 2019
We
evaluated
SpCas9
activities
at
12,832
target
sequences
using
a
high-throughput
approach
based
on
human
cell
library
containing
single-guide
RNA-encoding
and
sequence
pairs.
Deep
learning-based
training
this
large
dataset
of
SpCas9-induced
indel
frequencies
led
to
the
development
activity-predicting
model
named
DeepSpCas9.
When
tested
against
independently
generated
datasets
(our
own
those
published
by
other
groups),
DeepSpCas9
showed
high
generalization
performance.
is
available
http://deepcrispr.info/DeepSpCas9.
