►
From YouTube: DevoWorm (2023, Meeting #12): DevoLearn docs, Multiphysics, Bio + Physics Morphogenetic Curvatures
Description
Update of devoworm.github.io and DevoLearn documentation. More detailed discussion of APS March meeting. Multiphysics modeling of the embryo and finite element alternatives. Articles on the potential of modern Biophysics as an interdisciplinary enterprise and morphogenetic curvatures. Attendees: Richard Gordon, Sushmanth Reddy Mereddy, Susan Crawford-Young, Morgan Hough, and Bradly Alicea
A
B
D
Actually,
I
haven't
done
alone
this
all,
but
actually
another
guy.
Is
that
like
he
want
to?
He
was
a
web
guy
actually
and
he's
not
he's
so
interested
in
open
source
and
results.
So
we
completely
did
this
whole
work
and
the
idea
was
also
his
like
and
before,
when
I
was
started,
contributing
to
the
whole
Army.
D
D
D
Through
demographic
documentation-
and
these
are
like
what
have
been
done
from
2020
to
22-
actually
double
and
start
I
thought
of
started
in
2020
and
I
created
20
20
to
2022-
and
this
is
like
last
page
introduction
about
mentor
and
students
who
are
participated
in
this
use
of
on
this
work.
So
that's
and
I
have
sent
this
whole
thing
to
my
YouTube
app
actually
to
review
it.
Don't
like
yeah,
that's
the
main
problem
like
so
many
people
are
not
understanding
about
devoland
or
demograph
when
they
came
up
to
see.
D
So
he
give
me
some
kind
of
good
idea
to
create
another
blog
option
also
for
them
and
in
the
blog
option,
see
like
why
you
can
minor
really
did
a
pretty
good
work
on
this
thing
see
everything
is
like
clearly
documented
in
there.
This
thing,
Community
bonding,
is
all
all
the
these
all
links
are
there
for
their
work,
what
we
call
their
block
Pages.
These
are
Links
of
their
block
Pages
coding
period
week
from
week
to
week
three
weeks,
yeah
everything
is
documented,
but
many
people
don't
know
like
these
things.
Are
there
so
yeah.
B
C
D
C
D
C
D
Pages
everything
is
clearly
explained.
We
have
a
lack
of
documentation
from
demograph.
I
will
take
care
of
it,
like
I
will
try
to
add
some
tutorials
kind
of
thing
on
at
these
Pages.
If
we
integrate
into
this
main
website,
it
will
be
like
a
very
useful
tool
and
you
can
like,
if
anyone
comes,
we
can
directly
give
this
webpage
and
tell
them
to
explore
the
code
or
tools
or
even
like
how
they
are
built,
because
minex
myox
work
is
like
pretty
awesome.
He
mentioned
everything
how
you
trained
the
model.
D
D
D
D
You
can
handle
this
website
we'll
try
to
maintain
as
food
as
it
possible.
Okay,
if
anyone
who
just
want
to
write
about
blog
just
create
a
PR
write,
a
readme
file
like
what
they
wanted
if
they
keep
a
full
request.
If
it's
good
enough,
we'll
try
to
add
changes,
it's
not
good
enough.
It
will
not
merge
it
in
the
main
branch.
That's
what
we
are
thinking
actually.
B
C
B
A
D
C
D
D
Actually,
this
book
I
was
reading
for
my
exam
tomorrow.
I
have
an
exam
related
to
ml.
This
is
the
same
book.
I
need
to
start
it
now
there
is
a
midterm.
Examination
is
going
on.
This
is
the
same
book.
Thank
you,
yeah!
That's,
that's
all
my
updates,
and
for
next
week
I
was
thinking
to
work
on
this
hugging
face.
D
D
D
C
B
All
right,
that's
great.
Thank
you.
Thank
you
for
the
update,
yeah
and
glad
to
see
that
we're
working
on
this.
So
that
was
something
the
github.io
is
something
we've
been
building
for
a
couple
years
now
we
in
2020,
when
we
did
Evo,
learn
we
we
kind
of
had
all
the
models.
It
was
kind
of
a
place
where
we
collected
a
lot
of
the
models
and
then
put
them
in
that
place.
That
was
actually
before
we
really
started.
B
The
diva
learn
trajectory,
so
it
was
pretty
out
of
date
but
I
like
the
idea
of
having
you
know,
updating
it
to
where
we
are
with
Diva,
learn
and
evil
graph
and
then
having
like
these
tutorials
available,
so
people
will
know
kind
of
they
have
you
know
a
lot
of
software
has
like
tips
and
tricks,
or
they
have
FAQs
or
something.
This
I
think
is,
is
a
good
way
to
do
that
so
yeah
and
then
we
can,
you
know
continue.
B
I
would
recommend
that
for
people
doing
gsoc
that
we
do
block
weekly
blog
posts
to
keep
updated
incf
used
to
require
that
for
gsoc,
but
last
year,
I
don't
think
they
did
so.
We
didn't
do
it,
but
you
know
I,
think
that's
a
great
idea
so
keeps
keeps
everything
you
know
keeps
everything
you
do
in
context.
It's
like.
B
D
E
C
E
E
Well,
they
take
spheres
and
pack
them
and
they
end
up
with
hexagons
on
a
flat
surface.
And
then,
if
they're
on
a
curved
surface,
then
there's
gaps
in
in
it
and
then
they've
got
something
that
looks
like
a
oblong
that
looks
like
a
fly
embryo
that
they're
working
with
it's
quite
interesting
anyway,
thanks
they
use
Marvel
and
I'm,
hoping
that
that
might
be
somewhere
between
console
and
Matlab.
Somehow
I
can
get
a
model
in
Matlab
and
I
can
Port
it
into
console.
E
Yeah,
including
the
fact
that
the
elasticity
in
the
actin
actin
filaments,
is
quite
stiff,
so
you
get
a
J
curve
with
yeah
at
the
tensegrity
when,
when
the
acting
action
or
elastic
filaments
or
the
strings
in
your
model
are
quite
stiff,
it
seems
to
work
that
way.
I
have
a
paper
about
someone
who
did
that
with
with
the
Triangular
tensegrity
and
that's
the
way
it
worked
and
I
think
I
presented
that
yeah.
E
A
E
Also,
in
the
aps,
physics,
they
were
going
about
hydrolonic,
acid
and
brush
borders
and
how
entergen
just
attaches
at
certain
places
and
not
all
over
the
place
in
a
Cell,
because
you've
got
these
brush
quarters
and
I
guess
they're,
also,
maybe
along
the
sides.
So
the
attachments
really
are
just
at
the
corners
of
yourself
whatever
well,
whatever
Corners
the
self
comes
up
with
they
change.
But
what.
E
Yeah
I
I
actually
captured
the
guy's
lecture.
I
went
they
had
they
have
them
the
lectures
after
the
conference.
They
have
some
of
the
main
ones,
and
this
this
fellow's
lecture
was
there
and
so
I
captured
a
couple
of
images
he
didn't
want
it
recorded.
So
I
was
careful,
yeah
yeah
that
I
could
show
you
I
mean
I
could
show
you
what
he
means,
but
but
it
was
it's
nice,
it's
basically
there's
just
these
little
brushes
that
grow.
C
C
C
B
A
E
Matrices
produced
by
the
finite
element,
analysis
and
when
it
they
also
want
them
to
it.
To
say
why
that
all
I'm
getting
out
of
console
is
an
empty
matrices,
more
full
yeah,
actually
Timothy,
Alberni
I
think
it
is
from
tops
said
he
would
try
it.
He
said
he
said
he
thought
my
model
would
work.
Yeah
I
want
to
download,
Morpho
and
use
it
now.
I
was
like
tired
of
someone's
other
stuff,
but
anyway,
I
get
to
try
Matlab
this
week.
E
E
I
actually
there's
a
book
written
by
Gan.
The
guy's
name
is
last
name
is
Dan
and
he
has
pre-code.
He
has
code
that
I
can
use
actually
I've
used
it
and
made
my
model
and
and
not
that.
But
it's
the
the
Twisted
model,
the
Standalone
twisted
Integrity,
so
I
just
have
to
straighten
it
out
and
put
some
glips
on
the
side
and
I
had
a
layer.
A
B
E
B
Right
so
yeah,
Morgan
and
I
actually
had
a
nice
talk
this
week
about
multi-physics
models
and
finance
element,
analysis
and
kind
of
the
things
that
he
found.
You
know
he's
finding
a
lot
of
the
stuff
we're
talking
about
interesting
and
he
was
inquiring
about
that.
So
and
I
I
recalled
that
we
talked
about
finite
element
models
like
very
early
on
in
Diva
worm.
Dick
remembers
like
2014
2015.
We
talked
about
that.
B
A
Yeah
I
think
that
the
most
fascinating
part
of
that
is
Wayne,
Broadway
and
stuff.
We're
using
Finance
doing
the
inverse
sliding
element
model
where
he
could
use
fine
allowance
to
measure
the
forces
between
cells
and
embryos.
Yeah,
okay
and
he's
retired
he's
gone
off
and
he's
not
doing
anything
with
science
so
like
it
would
require
resurrecting
that
stuff
from
the
papers.
B
F
E
What
they
want
to
do
in
with
what
I'm
working
on
is
inverse.
They
want
to
do
the
inverse
method
with
it.
So
I'm
with
the
tensegrity
you've
got
something
in
sin
compression.
It
has
one
value
and
something
that's
the
intention.
It
has
another
value,
and
so
you
can
more
easily
do
an
inverse
method
with
two
just
two
unknowns:
foreign.
A
F
Yeah
yeah,
so
we
we
were
doing
a
skull
conductivity
estimation
by
by
injecting
current
into
the
heads.
That
was
that's
what
I
was
doing
or
that's
what
scientists
were
doing
at
electrical
geodesics
for
these
head
models,
but
but
I'd
be
you
know
and
yeah.
It
does
sound
like
like
console's,
not
working
right
if
you're
getting
empty
matrices.
E
Well,
that's
only
if
I
my
model
is
not
working,
I
can
get
it
to
work,
but
trying
to
figure
out
why
it's
not
working
like
I
can
I
can
take
it
to
the
edge
of
where
it
works
and
I
actually
have
one
that
one
set
of
variables.
That
sometimes
gives
me
a
result
and
sometimes
gives
me
an
empty
Matrix
just
depending
on
maybe
the
which
way
I
which
hand
I
use
to
push
the
button.
E
E
A
E
E
E
E
C
C
E
A
B
E
C
F
F
You
know
five
models,
but
there
are
some
packages
for
it
specifically
in
you
know
in
my
field
that
like
build
on
top
of
packages
and
others
that
that
are
commonly
used,
you
know
like
it's
sort
of
like
an
alternative
to
Phoenix,
but
but
yeah,
but
very
interested
and
and
the
tense
gracity
in
particular.
You
know
I'd
really
like
to
see
what
how
to
how
to
try
and
fit
that
into
a
model.
E
I'm
really
interested
in
trying
to
transfer
this
knowledge
to
you
know,
that's
more
like
cellular
work,
yeah
the
it's
just
a
matter
of
building
a
tensegrity
and
it's
just
tension
Elements,
which
are
basically
strings
and
depression
Elements,
which
are
rods,
and
you
just
make
the
rods
fall
apart
in
tension
like
they
just
do.
Compression
and
the
tension
elements
just
do
tension
like
they're
a
rope.
You
can't
push
a
rope
so.
E
A
E
C
A
A
E
F
B
E
A
We'll
do
this
paper
on
epithelia
as
bubble
wraps,
and
we
just
we
only
examined
real
bubble
wraps.
In
other
words,
we
made
bubbles
between
two
glass
plates,
okay
and
then
measure
their
curvatures,
and
that's
that's
the
paper
from
which
Wayne
developed
this
technique.
You
can
generalize
it.
We
were
doing.
We
did
a
very
simple
case
and
we
generalized
it
to
follow.
Epithelia.
E
A
A
E
A
E
E
B
So
I'm
going
to
share
my
screen
and
talk
about
some
items
of
Interest.
This
is
an
interesting
piece
that
just
came
out
in
EPS
news,
so
we're
talking
about
APS
again.
This
one
is
an
opinion
piece
by
William
Bialik
who's,
a
big
name
in
in
biological
physics
and
so
he's
reflecting
on
biological
physics
Coming
of
Age.
So
this
is
brand
new
and
he's
basically
recounting
you
know
how
biological
physics
started
and
how
it's
now
coming
of
age.
So.
A
B
I,
don't
know
we'll
see
in
a
minute
here,
yeah
once
an
awkward
confrontation
between
disciplines.
Biological
physics
is
having
its
moment
and
showing
that
life
is
not
just
a
mess,
so
he
kind
of
goes
back
to
the
18th
century.
For
some
of
the
origins
here.
So
physics
and
biology
were
not
always
separate
disciplines.
An
18th
century
controversy
about
animal
electricity
were
foundational
for
the
understanding
of
electricity
more
generally,
so
this
is
galvani
and
his
experiments
with
muscle
and
electricity
and
even
Durham
I
think
did
some
experiments
of
electricity
and
muscles.
B
So
that
was
something
that
people
were
really
interested
in
at
the
time
and
it's
you
know
it,
but
that's
something
that
before
physics
and
biology
really
became
like
it
came
into
their
own
as
disciplines,
I
guess
in
the
19th
century,
explorations
of
vision
and
hearing
intermingled
with
the
emerging
understanding
of
Optics
and
Acoustics.
So
that's
that's
another
kind
of
point
of
intersection
there.
B
B
So,
and
still
today,
even
you
know
a
lot
of
the
work
on
some
of
the
tools
for
genomics
are
done.
You
know
in
like
maybe
a
biochemistry
department
or
somewhere
else.
You
have
a
computer
science
departments,
even
so
that's
kind
of
interesting
how
these
things
migrate
to
different
fields,
and
you
know,
provide
these
opportunities
for
interaction
between
the
fields
or
between
departments.
B
B
Physics
was
the
other
way.
The
style
of
thinking
was
attractive
and
the
theory
is
elegant
and
Powerful,
but
I
never
had
an
original
idea
about
problems
in
the
fields
traditional
core,
so
yeah
I
mean
that's
true,
I.
Think
in
in
a
lot
of
fields
where
people
have
this
sort
of
urge
to
be
interdisciplinary
because
of
the
maybe
the
problems
they're
interested
in
you
know
they
don't
necessarily
fit
into
the
core
of
the
field,
but
also
you
know.
Sometimes
a
field
like
you
know
biology.
They
don't
really
talk
like
I'm
thinking
of
computational
biology.
B
For
me,
you
know
in
in
a
traditional
biology
department.
They
don't
talk
about
the
problems
in
a
way,
that's
satisfying
because
you're
always
thinking
of
computational
aspects
of
the
problem,
whereas
you
know
in
computer
science,
for
example,
you
don't
usually
the
idea
of
you
know
biological
computation
or
you
know,
biology
or
computation
for
biology.
Isn't
that
the
center
of
the
field?
It's
kind
of
like
this
thing
on
the
edge
of
the
field
and
so
you're
going
through
all
the
stuff
that
you
find
extraneous
to
get
to
the
thing
you're
interested
in
so
yeah.
B
I
I
have
a
similar
experience
to
this,
but
with
computer
science,
and
so
it
was
clear
that
physicists
were
doing
all
sorts
of
interesting
things
connected
to
the
living
world.
But
these
efforts
didn't
coherent
to
a
community
and
certainly
not
into
a
recognizable
branch
of
physics.
We
went
to
meetings
where,
mostly,
we
would
find
biologists
working
on
the
same
systems,
but
not
physicists,
asking
the
same
kinds
of
questions.
B
If
we
asked
our
biology
professors,
they
would
argue
that
a
physicist
search
for
Simplicity
and
universality
was
an
obvious
conflict
with
the
complexity
and
diversity
of
life.
If
we
asked
their
physics
professors,
they
might
talk
about,
a
colleague
who'd
become
a
biologist
and
perhaps
learn
that
experiments
in
biology
are
messy.
B
So
this
is
kind
of
like
when,
before
a
really
biological
physics
was
a
quote-unquote
thing,
and
so
you
know
Community
kind
of
was
people
kind
of
moving
between
fields
or
you
know
whatever
and
then
I
guess.
Now
it's
becoming
sort
of
a
Newfound
relevance,
I
guess
so
40
years
later,
which
was
I
guess
now
or
in
the
recent
past,
the
animation
landscape
has
changed
radically.
B
E
B
So
that
that
puts
in
the
chat
given
a
spherical
Kyle,
the
physicist
view
of
life.
So
that's
a
con.
That's
a
yeah
problem.
We've
talked
about
a
lot
in
the
meetings,
the
spherical
cow,
which
of
course,
is
where
there's
a
joke
about
physicist
modeling,
something
you
know
in
biology
and
they
say
assume
a
sphere
in
a
vacuum
and
that's
always
like
the
first
step
to
modeling.
B
B
B
So
you
know
the
message
of
biological
physics
is
a
beautiful
phenomena
of
Life
connected
deep
physical
principles,
so
things
like
bacterial
swimming
and
sensing,
you
can
view
those
as
being
dominated
by
low
rentals
number
of
mechanics
and
diffusion,
and
that
gives
you
a
new
insight
into
these
behaviors.
We've
talked
a
lot
about
that
in
the
meetings
here
where
you
know
we're
talking
about
behaviors
and
the
physics
of
cells,
the
physics
of
in
a
single
cell
organisms,
vision
and
photosynthesis
illustrate
how
quantum
mechanics
can
produce
a
broad
absorption
band
rather
than
narrow
spectral
lines.
B
So
in
photosynthesis,
people
are
interested
in
Quantum
effects
and
envision.
You
know
there's
a
lot
of
like
physics
and
psychophysics
and
and
regular
physics,
and
then
DNA
exemplifies
the
random
polymer
that
appears
in
all
statistical
mechanics
courses.
So
a
lot
of
like
molecular
diffusion
and
a
lot
of
things
that
involve
you
know
biochemistry
have
this
sort
of
statistical,
mechanics,
flavor
and
we've
talked
about
statistical
mechanics
a
bit
more
broadly
than
that,
even
in
this
group.
So
that's
something
that
is
now
getting
some
more
attention.
B
B
Is
there
a
mechanical
engineering
department
and
it's
a
new
view
on
how
tissue
flows
in
the
embryo,
so
this
is
Columbia
Engineers,
develop,
novel
technique,
understand
the
forces
that
control
cell
shapes
and
tissue
flows
androsophil
embryos,
and
so
this
is
an
example
of
a
so
in
the
drosophil
embryo.
You
get
this
huge
centitium
full
of
it's
a
single
layer
and
it
has
a
bunch
of
nuclei
in
it.
It's
multi-nucleated
and
then
that's
something.
They
call
Synthesia
and
I'm,
not
sentitious
cellularizes
and
becomes
all
these
cells.
B
So
if
we
had
another
message
here,
interest
Morgan
said
interesting:
Susan
cited
the
person
Timothy
Atherton
okay,
so
this
is
on
a
new
paper
that
was
published.
So
basically
they
want
to
know
whether
we
want
might
thought
we
might
be
able
to
learn
something
about
whether
the
embryonic
tissues
are
solid
or
fluid.
Just
by
looking
at
the
shapes
of
cells
in
the
tissue.
So
I
guess
this
means
that
they
want
to
know
if
it
behaves
like
a
solid
or
a
fluid.
B
B
So
this
is,
you
know
again
with
you,
have
to
use
traditional
engineering
approaches,
but
you
want
to
be
able
to
measure
these
mechanical
properties
and
see
what
tissues
behave
like
solids,
maintaining
their
shape
and
resisting
flow
in
which
tissues
behave
like
fluids
flowing
usually
easily
and
changing
shape.
So
you
have
these
flow
fields
in
the
embryo.
B
You
have
cells
that
are
sort
of
compliant
to
those
flows
or
cells
that
resist
those
flows,
and
so
in
this,
in
this
study,
they
focused
on
a
very
fast-moving
developmental
event,
which
the
embryonic
tissue
rapidly
changes
shape
to
elongate
the
head
to
tail
body
axis
of
the
fly.
So
these
drosophil
embryos
are
long,
they're,
oblong,
very
oblong,
actually
for
an
embryo,
and
they
have
this
polarity
early
on
and
these
flows
go
from
anterior
to
posterior.
So
we
have
this:
let's
have
a
little
animation
of
these
dramatic
tissue
flows.
B
B
There
I
think
both
I'm
not
really
sure
how
the
force
is
generated,
but
there's
it
I
think
has
something
to
do
with
like
the
sort
of
the
shape
of
the
embryo
or
like
there's
some
cells
that
are
making
you
know
making
forces
and
then
those
are
transmitted
across
the
embryo
I'm,
not
really
sure
exactly
what
the
mechanism
is.
But
you
know
Susan.
E
E
E
E
This
follows
through
to
the
this
any
structure
you
make
right
of
a
cell,
it's
a
hexagon,
and
how
can
that
change
from
the
hexagon?
Well,
one
way
is
you
just
have
different
forces
holding
up
the
activities
and
nodes
on
the
corners?
That's
one
way
you
can
do
it,
and
but
the
cells
also
eliminates
their
sides
like
instead
of
a
hexagon,
they'll,
think
of
a
pentagon
and
then
a
square
like
it.
Just
they
regulate
themselves
right.
F
E
B
Right
so
yeah
we
actually
last
week,
I
did
a
thing
I,
do
it
after
the
end
of
the
official
meeting
and
I
cover
some
papers
on
the
YouTube
channel,
so
those
are
something
you
know.
I
want
to
check
out
at
the
end
of
the
last
recording
you
know,
with
with
the
papers
were,
we
did.
I
did
one
a
couple
on
curvatures
and
condenses,
and
basically
it's
this
idea
of
curvatures
and
cells,
and
oh
there
we
go.
There's
the.
E
E
B
B
C
B
So
yeah
last
week,
I
did
some
papers
on
curvatures
and
condensates,
so
they
were.
People
were
looking
at
different
macromolecules
in
a
in
a
condensate
like
a
they
were
using
a
water
droplet
and
they
were
looking
at
some
of
the
organization
of
things
in
there
and
then
they
were
using
looking
at
drosophila
embryos
and
looking
at
this
problem
of
cellularization
and
then
these
fluid
flows.
So
this
is
the
common
problem
of
what
happens
to
these
things
in
a
flow.
Do
you
get
like
you
know?
B
Actually,
I
think
they
were
also
using
bubbles
in
this
condensates
paper,
so
they
were
looking
at
bubble
formation
bubble
transport
and
then
they
were
then
the
other
paper.
They
were
looking
at
cellular
movements
in
the
drosophila
embryo,
so
this
week
I
wanted
to
talk
about
two
other
papers,
and
these
are
basically
drosophila
embryo
specific.
B
So
this
is
the
first
paper
anisotropylink
cell
shapes
to
tissue
flow
during
convergent
extensions.
This
is
what
Susan
was
talking
about
with
convergent
extension.
They
don't
talk
about
tensegrity
models
in
this
paper,
but
they're
talking
about
this
idea
of
tissue
flows
and
cell
shapes
and
convergent
extension.
B
So
the
abstract
here
reads
within
developing
embryos,
tissue
flow
and
reorganize
tissues,
flow
and
reorganized
dramatically
on
time
scales
as
short
as
minutes.
This
includes
epithelial
tissues,
which
often
narrow
are
often
narrow
and
elongate,
and
converge
and
extension
movements
due
to
anisotropies
and
external
forces
or
internal
cell
generated
forces.
So
this
is
where
they're
elongating
according
to
different
types
of
forces
and
different
orientations,
and
so
you
know
that's
kind
of
like
you
know,
some
set
of
forces
or
Force
interactions
will
trigger
this.
A
B
However,
the
mechanisms
that
allow
or
prevent
tissue
reorganization,
especially
in
the
presence
of
strongly
anisotropic
forces,
remain
unclear.
So
this
is
something
they
don't
really.
This
is
what
they're
going
to
answer
here,
but
it's
probably
Still
Remains
unclear,
we'll
see
we
study
the
question,
this
question
in
the
converging
and
extending
drosophila
germ
band
epithelium,
so
there's
this
germ
band
actually
I
have
some
jamboard
images
here
of
this.
So
this
is
a
where
you
get
this
cell
flow.
B
You
get
convergent
extension
and
you
get
these
flows
that
move
cells
around
the
embryo,
I,
don't
actually
know
if
I
have
an
image
of
this,
but
okay
I,
don't
basically
it's
where
you
have
this,
the
drosophila
embryo
and
you
have
the
furrow
moving
across
it.
We've
talked
about
the
furrow
before
and
you
have.
B
So
this
is
the
drosophil
embryo,
the
cellularization,
and
you
have
this
Furrow
moving
across
it,
and
you
have
germ
bands
that
form
in
this
structure
and
so
you're
starting
to
get
you
know,
structure
moving
across
the
embryo,
and
so
this
is
just
the
context
for
this.
Where
you're
going
to
see
these
cell
flows,
you
see
them
at
the
edge
of
the
embryo,
but
you
also
see
them
inside
the
embryo
and
it's
going
to
start
moving
things
around.
B
So
this
is
so.
We
studied
this
question
the
converging
and
extending
drosophila
German
band
epithelium,
which
displays
planar,
polarized,
myosin,
II
and
experiences
an
esotropic
forces
from
neighboring
tissues.
We
show
that,
in
contrast,
isotropic
tissues
cell
shape
alone
is
not
sufficient
to
predict
the
onset
of
Rapid
cell
rearrangement,
so
their
first
testis
of
cell
shape
and
they
figure
out
that
it's
not
sufficient
to
predict
this
from
theoretical
consideration
and
vertex
model
simulations.
So
this
is
where
they're
doing
some
simulations.
B
We
predict
that
in
isotropic
tissues
to
experimentally,
accessible
metrics
of
cell
patterns,
the
cell
shape
index
and
a
cell
alignment
index
are
required
to
determine
whether
in
an
isotropic
tissue
is
in
a
solid
lake
or
fluid-like
state.
So
they
basically
want
to
know
if
something
is
in
a
solid-like
or
fluid
like
they
mentioned
in
the
preview
article.
You
know
they
have
to
determine
the
physics
at
a
sort
of
the
state
of
the
tissue.
B
We
show
the
changes
in
cell
shape
and
Alignment
over
time
and
The
drosophila
Germ
band
predict
the
onset
of
Rapid
cell
rearrangements
in
both
wild
type
and
snail
twist
mutant
embryos,
so
the
wild
type
are
the
ones
that
are
normally
developing
snail
Twist
of
some
mutant
phenotype,
and
so
you
can
look
at
this
in
different
ways.
You
know
where
there
may
be
differences
in
the
process.
B
So
then,
where
our
theoretical
predictions
is
further
are
further
improved
when
we
also
account
for
cell
packing
disorder.
So
this
is
where
there's
disorder
in
the
cell
packing.
These
findings
suggest
that
conversion
extension
is
associated
with
a
transition
to
a
more
fluid-like
tissue
Behavior,
which
may
help
accommodate
tissue
shape
changes
during
rapid
developmental
events.
B
So
this
goes,
you
know.
If
you
go
down
to
the
figures,
they
show
this
germ
band
epithelium
here
this
is
the
embryo.
You
have
these
different
part.
This
is
the
part
that
they're
highlighting
in
the
Box.
This
shows
The
Germ
band
epithelium
down
care,
so
that
would
be
down
in
this
region
here
and
then
this
shows
a
little
square
of
that
epithelium.
So
you
have
an
internal
stress
from
a
planar
polarized,
myosin
II.
So
this
these
are
myosins
in
here
and
external
stress
from
neighboring
tissues.
B
E
B
B
They
rearrange
so
that
you
have
these
cells
touching
these
top
and
bottom
cells,
touching
each
other
and
the
middles
in
the
the
cells
in
the
middle
here
in
this
middle
band
are
forced
outward
that
cell
rearrangement,
it's
basically
where
they're
pushing
together
and
they
push
each
other
out
of
the
way
in
the
area
range
like
that,
then
there's
cell,
stretching
where
you
have
these
the
cell
packing
that
stretches
in
shape.
So
it's
just
stretching,
but
it's
not
rearranging
in
any
way,
and
so
they
show
here.
B
This
is
access
elongation.
Over
time
you
have
a
relative
anterior,
posterior
tissue
length
and
cell
rearrangement
rate,
which
is
this
relative
AP
cell
link
change.
So
you
have
two
y
axes
and
I
think
that
why
I
don't
know
the
purple
and
the
blue,
so
the
purple
here
is
celery
Arrangement
rate
and
the
green
is
relative.
B
Ap's
sublink
change,
so
they're
different
y-axis
per
color,
and
then
this
one
is
those
are
equivalent
to
relative
AP
tissue
length,
and
so
basically
the
bottom
line
of
this
is
that
over
time,
the
interior
posterior
cell
length
change
decreases
when
we
get
towards
zero
and
then
increases
after
I
guess.
This
is
the
event
that
they're
measuring
before
the
event
you
have
this.
B
It
kind
of
goes
to
zero
kind
of
goes
below
zero
at
the
time
of
the
measurement
and
then
up,
and
then
it
settles
back
down
to
zero
after
like
about
30
30
minutes
after
so
this
whole
event
takes
place
in
this
period
of
time.
The
cell
rearrangement
rate,
however,
stays
at
zero
until
about
five
minutes
after
the
event
and
then
goes
up,
there's
a
peak
around
10
minutes,
and
this
is
of
course,
probably
where
you're
getting
a
lot
of
this
flow
reactivity
to
the
flow.
B
So
there's
another
figure
here
where
they
kind
of
show
this
number
of
cell
sides
vertex
coordination
number
and
they
do
these
vertex
model
simulations.
So
this
is
where
you
get
different
fractions
of
different
cell
types,
so
you
get
five
six
and
seven
you
get
this
fraction
of
pentagonal
cells
on
this
axis.
You
get
the
average
shape
Index
P
bar
on
this
axis,
and
you
can
see
that
they're.
B
You
know
they're
simulating
a
packing
of
cells
and
they're,
showing
these
simulations,
so
they
have
the
fraction
of
fluid
tissue
States
going
from
zero
to
ones
as
near
zero
is
blue
near
one
is
red
and
you
can
see
that
they
go.
It
goes
above
this
regression
line
in
red
and
it
goes
below
the
progression
line
in
blue.
So
there's
this
fraction
of
fluid
tissue
States,
that's
lower.
B
You
know,
as
we
go
below
this
trend
line
in
higher
as
we
go
above,
which
is
kind
of
interesting.
If
you
look
at
this
for
five-sided
cells,
you
see
this.
The
blue
or
down
here
and
the
red
are
up
here
and
then
the
vertex
coordination,
the
router
up
here
and
the
blue
are
down
here.
So
this
is
kind
of
interesting
I.
B
Don't
know
what
that
means
exactly,
let's
see
what
they
say
in
the
in
the
legend
cell
shape
and
packing
disorder
alone
are
not
sufficient
to
predict
the
onset
of
cell
Arrangements
in
The
Germ
band
confocal
images
from
time
likes
movies
of
epithelial
cell
patterns
in
the
ventral
lateral
region
of
the
germ
band
tissue.
That's
what
we're
looking
at
here
or
we're
analyzing
cell
outlines
we're
visualized
by
using
the
fluorescent
latex
cell
membrane,
marker
Gap,
43,
M
cherry,
so
this
is.
B
B
B
B
The
dashed
line
is
the
prediction
from
this
reference
here:
instantaneous
cell
rearrangement
rate
per
cell
in
the
tissue
is
represented
by
the
color
of
each
point
up
here,
with
blue,
indicating
low
rearrangement
rates
and
run
to
Yellow
indicating
the
high
rearrangement
rates.
So,
basically,
you
have
low
rearrangement
rates
versus
high
rear
Arrangement
rates
and
I
find
that
if
you
look
at
the
shape
index
that
predicts
these
rearrangement
rates
for
different
size
cells
and
different.
B
So
this
is
these:
are
the
pentagons
here,
the
vertex
coordination
numbers,
where
you
have
how
many
vertices
that
the
when
the
cells
are
in
a
packing,
how
many?
What
what
size
vertices
vertex
forms.
If
you
have
a
three
cells
that
are
coming
together,
it's
three:
if
you
have
four
cells
that
are
coming
together,
it's
four
and
then
five
cells
coming
together,
it's
five
and
I
guess
the
implication
of
that
is
that
you
get
more
opportunities
for
these
forces
to
generate
a
flow.
B
B
B
B
Yeah,
so
this
is
the
vertex
model
and
then
this
is
increasing
strain,
so
you
have
this
box
of
or
this
field
of
cells
and
then
you're
compressing
it
as
we
go
down.
So
this
is
increasing
strain
and
then
this
is
the
effects
of
strain
on
model
tissue.
So
this
is
this
average
shape,
Index,
P
bar
and
then
against
the
shear
modulus,
which
is
G
so
now
we're
looking
at
Shear
forces
and
we're
looking
at
this
increasing
strain.
B
As
we
look
at
these
different
points
here,
I
guess
these
are
different
cells
in
these
packings.
So
this
is
the
blue
packing
here.
This
is
the
share
modulus
for
a
wide
range
of
values
and
then
the
average
shape
index
is
low
as
we
get
increase
the
strain
in
this
case,
the
purple.
We
see
that
it's
for
a
lower
range
of
share,
module,
I
and
then
it's
it's
a
higher
shape
index
and
then
for
the
red,
which
is
this
tightly
strained
or
highly
strained
packing.
You
see
these
red
points,
which
are
a.
C
B
Bit
less
sheer
modulus,
maybe
about
you,
know
a
half
of
the
range
from
zero
to
one,
but
you
get
an
even
higher
shape
index,
and
so
this
is
where
you
know
this
is
kind
of
an
interesting
result.
Again,
I,
don't
think
this
is
simulated.
This
is
just
demonstrating
what's
going
on
and
then
you
get
this
cell
shape
alignment
or
Q
versus
average
shape
index
or
P
bar,
and
so
you
can
make
predictions
for
different
values
of
these
parameters
so
for
high
P
bar
low
Q.
B
B
I
think
this
is
like
a
lot
of
looks
like
a
lot
of
three
and
four
vertex
points
in
in
this
Matrix
and
this
Matrix.
You
have
a
number
of
five
five
vertices
and
then
here
you
have.
It
almost
looks
like
an
ordered
packing,
but
I
don't
know
with
if
there's
any
characteristic
number
of
vertices
and
then
this
this
area
down
here,
which
is
I,
guess
High,
P
or
low
op
bar
High
Q.
That's
actually
geometrically
impossible.
B
So
that's
an
interesting
kind
of
way
to
unpack
some
of
this
note
one
intended
it's
just
showing
these
different
packing
possibilities
and
what
they
look
like.
So
that's
and
then
this
is
external
deformation.
So
this
is
where
you're
forming
this
packing,
and
this
is
using
cell
alignment,
shape
Q
versus
average
shape
Index
P
bar
this
is
showing
sort
of
under
external
deformation.
The
same
result,
in
figure
two
and
the
same
thing
for
internal
stress
which
stretches
this
packing
outward,
and
then
you
get
the
same
result
for
Q
versus
P
bar.
E
B
B
So
I
just
want
to
talk
about
this
paper
briefly,
this
is
a
nature
physics,
paper,
Collective
curvature,
sensing
and
fluidity
and
three-dimensional
multicellular
systems.
So
this
one
is
about
Collective
cell
migration,
and
this
is
I.
Don't
think
this
focus
is
as
much
in
a
drosophila,
but
let's
see
what
they
have
to
say
so
Collective
cell
migration
is
an
essential
process
throughout
the
lives
of
multicellular
organisms,
for
example,
in
embryonic
development
wound
healing
and
tumor
metastasis
substrates
or
interfaces
associated
with
these
processes
are
typically
curved.
B
So
there's
this
curvature
aspect,
you
know
that
we
talked
about
in
drosophila
embryos.
So
if
you
look
at
the
edge
of
the
embryo,
there
there's
a
curvature
and
that
curvature
has
some
implication
for
some
generating
some
of
these
forces,
especially
in
the
outside.
You
have
curvature
and
even
on
the
inside,
you
have
curvatures
once
the
tissues
are
formed.
So
you
get
these.
You
know
these
are
important
kind
of
structures
to
hang.
You
know
it's
not.
The
embryo
is
in
a
sphere.
B
There's
it's
a
sphere
of
maybe
sort
of
for
a
reason,
but
substrates
your
interfaces
associated
with
these
processes
are
typically
curved,
with
radii
of
curvature
comparable
to
many
cell
lengths.
So
there's
this
radii
of
curvature,
which
is
like
how
curved
is
it?
Is
it
very
tightly
curved
or
is
it
like?
You
know,
a
oblong
shape
and
so
using
both
artificial
geometries
and
lug
lung
alveoli
spheres
derived
from
Human
induced
pluripotent
stem
cells?
This
sounds
like
some
sort
of
organoid
work.
B
Actually
here
we
showed
that
cells
sends
multicellular
scale
curvature
and
then
it
plays
a
role
in
regulating
Collective
cell
migration.
So
this
is
where
we're
they're
arguing
that
cells
are
sensing.
This
curvature
they're
sense
in
the
curvature
of
this
microenvironment,
where
this,
the
embryo
or
whatever
it
is,
and
it's
playing
a
role
in
regulating
Collective
cell
migration.
B
So
you
can
imagine
that
the
cells,
you
know
kind
of
stop
at
the
boundary
or
they
kind
of
know
that,
like
they
don't
flow
outside
of
the
boundary
under
normal
conditions,
they
kind
of
find
a
boundary
and
it's
it's
usually
spherical
and
there's
a
curvature
to
it
where
you
know.
Otherwise,
we
see
in
some
of
these
other
in
these
hypothetical
examples
in
the
last
paper
where
they
use
the
square
packing
and
with
we
have
actually
in
biology,
is
a
curvature.
B
E
B
B
B
B
Aspect
to
it,
so
it's
it's
using
like
free
energy
and
minimizing
that
to
form
these
spheres
and
of
course,
we
know
from
like
Collective
material
or
the
collective
Behavior
materials
that
you
have
often
a
physical
constraints
to
like
what
the
shapes
that
they
form.
So,
if
you
put
you
know
like,
even
if
you
put
like
metal
balls
in
an
electric
field,
they
can
form
different
patterns
and
they,
but
they
tend
to
do
this,
not
because
they
know
what
they're
doing,
but
because
that's
the
free
energy
path.
C
B
So
we
observe
that
cluster
size
grows
linearly
as
sphere
radius
increases
which
further
stabilizes
the
multicellular
flow
field
and
increases
cell
collectivity.
As
a
result,
increasing
curvature
tends
to
promote
the
fluidity
in
multicellular
monoair
together.
These
findings
highlight
the
potential
for
a
fundamental
role
of
curvature
and
regulating
both
spatial
and
temporal
characteristics
of
multicellular
systems.
B
B
Curvature
can
greatly
change
ordering
classic
problem
acting
serious
as
tightly
as
possible
have
a
simple
solution:
a
flat
space
meaning
to
hexagonal
packings
and
2D,
and
hexagonal
plus
pack
pack
structures
in
3D
on
a
curved
surface
voids
in
the
packing
are
required
to
accommodate
the
curvature,
our
research
studies
and
structure
and
location
of
these
voids.
Now
they
are
affected
by
non-uniform
curvature
in
different
sized
particles,
that's
from
Tufts
from
the
Tufts
group,
I
guess
so.
B
So
let
me
see
if
we
have
any
figures
that
we
can
go
through
on
this.
So
this
is
this
shows
kind
of
the
curvature.
This
is
a
concrete,
an
a
it's,
a
concavity
and
b
it's
a
convexity
and
then
in
C.
You
can
see
that
this
packing
actually
shows
this,
so
this
might
be
like
a
tumor
and
another
tissue,
and
you
have
this
concavity,
the
convexity
just
showing
the
structure
of
this
sphere,
and
then
the
sphere
is
a
radius
and
an
angle
of
curvature,
which
is
here
and
that
angle.
B
Curvature
is
determined
by
the
physics,
but
also
it's
like
you
know.
The
cells
can
sense
it
and
they
kind
of
know
where
these
boundaries
are
once
it's
formed,
and
so
you
can
have
like
a
say
like
a
tumor
or
something
that's
spherical.
That
sort
of
you
know
the
cells
sort
of
form,
the
sphere
based
on
you
know
minimizing
their
free
energy
and
then
afterwards
they
can
sense
the
boundaries
of
this,
so
they
kind
of
have
a
guide
aside
from
free
energy
as
to
where
to
go
where
not
to
go.
So
this
is
the
nucleus
here.
B
These
are
empty
mkcd
cells
or
mdck
cells,
so
this
is,
these
are
grown
on
different
curvatures
and
they're,
individual
and
Collective
cellular
behaviors.
So
you
have
density
versus
velocity
and
you
can
see
this.
This
is
where
you
have
this,
this
value
for
k
at
different
sizes,
and
you
can
see
that
they
follow
this
this
trend
here,
and
so
this
is
the
sphere
here.
This
is
the
curvature
angle.
C
B
And
then
this
is
this
is
another
set
of,
so
this
is
the
multicellular
flow
field
and
for
MDC
K
cells
is
more
dynamical
with
a
larger
curvature.
So
this
is
a
Divergence
color
map
of
this
sphere
and
the
Divergence
values
along
the
surface
of
the
sphere.
This
is
an
example
of
the
sphere
with
its
angle
of
curvature
and
Divergence
values
across
this.
B
This
is
kind
of
showing
like
how
they're
they're
moving
around
in
this
hacking
or
and
then
yeah.
So
then
this
shows
sort
of
the
decrease
of
curvature
over
this
all
right.
So
this
is
this
is
for
the
alveoli
spheres,
and
so
these
are
derived
from
Human
disciplinary
potent
stem
cells.
This
confirms
a
collective
cellular.
Behavior
is
strongly
affected
by
curvature.
B
So
you
see
here
this
is
the
stain
of
this,
where
you
have
the
extracellular
Matrix,
you
have
these
these
cells
along
the
edge
and
then
the
curvature
decreases.
As
it
gets
larger
and
then
you
can
see
again
some
of
these
experiments
where
they
showed
the
Divergence
and
these
same
structures,
the
Divergence
gets
smaller
as
the
packing
gets
bigger.
B
B
A
C
E
E
Enough
with
the
sidelines
yeah,
okay,
yeah,
especially
the
the
second
one,
and
even
tough
University
yeah
yeah
stuff,
it's
it's
good
I'll,
just
quote
I'll,
just
quote
what
they've
gotten
yeah
they
they
have
an
interesting
oblong
embryo
at
that
there's
or
simulation
they're.
Studying
that
looks
like
a
fly
embryo
like
I,
said
earlier,
and
it's
there
somewhere
and
it's
a
misaligned
packing
in
the
in
this
oblong
embryo.
So
it
they
were
just
from
say
at
least
hexagonal
packings.