►
From YouTube: DevoWorm #38: Why no Lamarckism? Physics/shape in cell systems, Minimal cells, Light field imaging
Description
Simulating Lamarckism vs. Learning vs. Darwinian evolution and its connection to the acquisition of traits. Repeatability of developmental shape and tissue differentiation. The physics and shape of cells in disorders and cell populations. Self-organization of division rings in minimal cells (synthetic biology). Lightfield imaging. Attendees: Richard Gordon, Susan Crawford-Young, Bradly Alicea, and Jesse Parent
A
C
C
C
C
B
B
A
C
A
C
B
C
C
A
A
A
B
C
B
B
So
this
I
don't
know
if
I
can
zoom
in
on
this,
if
I
hide
that
probably
so,
this
is
a
picture
of
The
Hourglass
model
of
development.
So
this
is
where,
if
you
start
development
at
the
bottom,
you
have
your
eggs
and
they
start
to
differentiate
into
the
different
parts
of
the
mammalian
embryo
and
then,
as
they
move
forward
in
time,
you
start
to
get
what
they
call
Hawks
gene
expression
and
you
start
to
get
these
segments
of
the
embryo.
But
there's
this
period
called
the
phylotypic
period,
which
is
this
hourglass.
B
So
the
number
of
Pathways
that
the
embryos
can
take
is
they
you
know
is
the
animal
species
are
very,
you
know
somewhat
large
in
this
space
down
here,
but
then,
in
this
phyotypic
stage,
every
embryo
from,
like
you
know,
fishes
to
humans
to
mice
they
get
restricted
in
terms
of
how
they
express
their
Hox
genes
in
the
sort
of
the
form
of
the
embryos.
So
these
embryos
look
different
at
the
bottom.
B
They
have
very
diverse
Pathways
to
getting
to
this
stage,
which
is
the
phylotypic
stage
where
they
all
kind
of
look
similar,
and
they
have
this
sort
of
segmentation.
That
starts
to
occur
where
you
get
the
head
and
the
tail
and
the
midsection.
This
process
is
very
similar
and
then,
as
you
move
out
from
the
phylotypic
stage,
you
move
towards
different
phenotypes,
like
a
rat
and
a
rabbit
and
a
mouse
and
a
human
and
so
forth,
and
a
fish
and
so
there's
a
wide
variety
of
phenotypes
on
this
end
of
The
Hourglass.
B
B
A
A
B
A
B
Yeah,
so
this
yeah,
so
this
this
chart
actually
is
interesting
too.
This
shows
pre-gast
relation
early
gastrulation
and
this
shows
I
think
Mouse
embryo,
and
this
is
an
this-
is
comparing
like
like
what
they
call
a
gastroloid
versus
an
organoid.
So
this
is
an
organoid
when
they
have
the
flat
sheet
here
and
the
gastroloid
is
this
thing
that
has
shapes
to
it
and
so
they're
showing
that
you
can
get
these
different
segments
of
different
cell
types
and
you
get
movements
in
different
directions.
B
So
we're
talking
last
week
about
how
you
can
form
like
a
tube
that
curls
around,
and
this
just
shows
like
the
direction
of
cell
movement-
that's
kind
of
recapituated
in
some
of
the
some
of
the
embryoids
versus
in
like
a
something
that
approximates
an
actual
embryo.
So
this
is
where
you
have
pre
and
early
gastrulation
and
those
patterns
continue
and
they
kind
of
move
in
a
certain
direction.
B
You
see
that
in
check
as
well,
where
there's
movement
outward
in
a
flat
sheet,
but
in
this
thing
that
has
shape
it
moves
in
the
same
direction,
but
it's
a
little
bit
different
in
terms
of
its
shape,
and
so
you
have
these
different
things
that
you
see
in
different
embryos.
It's
it's
a
kind
of
an
interesting
figure.
I,
don't
know
what
to
make
of
it.
Beyond
this,
but
I
just
thought
this
was
an
interesting
figure
and
then
this
is
yeah.
B
A
B
B
Foreign,
so
I
mean
I
thought
that
figure
was
kind
of
interesting,
and
so
the
next
one
I
want
to
talk
about
is
this.
This
is
a
nice
article.
This
is
experimental
or
marchism,
and
this
is
out
of
American
scientists.
This
is
an
experiment.
This
is
Brian,
Hayes,
I,
don't
know
he's
right
here
for
American
scientists.
A
B
Yes,
would
this
evolutionary
strategy
worth
the
investment?
So
lamarckism,
of
course,
is
this
idea
from
Lamarck,
where,
if
you
have
say
like
an
animal
that
wants
to
sort
of
increase
its
neck
length
like
a
giraffe
or
wants
to
increase
its
muscle,
mass
or
something
it
just
behaves,
and
that
behavior
reinforces
increases
in
the
size
of
the
structure.
B
And
so
that's
how
well
Mark
proposed
that
the
giraffe
had
acquired
such
a
long
neck,
because
the
giraffe
went
to
eat
leaves
off
the
trees
and
it
kept
stretching
its
neck
upward
and
upward,
and
then
it
finally
ended
up
with
a
long
neck
and
of
course,
Darwin
knocked
that
idea
down,
because
first
of
all,
there
was
no
hereditary
reason
for
it
or
no
hereditary
mechanism
for
it.
But,
most
importantly,
it
didn't
seem
to
square
away
with
a
lot
of
the
evidence
that
you
saw
that
better
fit
was
better
explained
by
natural
selection.
B
But
you
know
people
have
used
this
idea
of
lamarchism
in
computational
modeling
in
genetic
algorithms,
they've
used
lamarchism
a
lot
to
sort
of
build
algorithms
that
do
certain
things.
So
the
reason
I
mentioned.
That
is
because
this
is
an
ex
a
computational
experiment,
comparing
well
markism
with
Darwinism
with
actual
learning
algorithms,
and
so
it's
an
interesting
thing.
Interesting
comparison,
so
you
know
lamarckian
ideas
existed,
they
were
pretty
darwinian
and
they
were
eventually
kind
of
became,
went
into
disrepute,
and
you
know
people
still.
B
This
people
know
what
it
is,
but
it's
like
what
is
it
really
explaining
so,
but
really
one
of
the
things
that
Lamarck
kind
of
focuses
on
is
this
idea
of
feedback
loops
and
so
the
trouble
with
right?
So
they
they
say
they
have
these
questions
about
lamarcki
and
lamarckism,
and
so
one
way
of
answering
them
is
to
note
that
molecular
Pathways
needed
to
further
marketing
inheritances
don't
exist
within
the
context
of
Life.
As
we
know
it,
there's
no
way
for
the
elephant's
nose
to
talk
to
the
elephant's
genes,
especially
the
germline
genes.
B
The
central
dogma
molecular
biologies
is
information
flows
from
DNA
to
RNA
to
protein,
not
the
other
way
around,
which
would
be
required
for
Lamarck
lamarckism.
A
lamarchy
and
feedback
loop
would
seem
to
require
some
mechanism
by
which
the
proteins
of
the
phenotype
could
alter
the
DNA
of
the
genotype.
So
but
then
that
invites
a
question
as
to.
Why
is
that?
Why
is
it
that
these
feedback
loops
never
evolved?
B
So
it's
not
impossible
that
those
feedback
loops
could
have
evolved
at
some
point,
but
they
never
did,
and
there
are
a
lot
of
shortcuts
and
genetic
detours
like
plasmids
transposons,
retroviruses
prions.
That
seems
a
bit
arbitrary
to
declare
this
one
pathway
out
of
bounds,
so
you
know
other
than
like
they're
obvious
things
about
the
central
dogma,
the
white
spin
barrier
that
hold,
but
why
have
we
not
seen
the
evolution
of
something
like
this
and
so
maybe
another
possible
reason
why
we
don't
see
little
marketing
inheritance?
Is
it's
not
worth
the
bother?
B
So
you
know
it's
just
not
worth
the
organism's
time
to
evolve.
This
pathway
and
I
mean
I
say
that,
like
as
sort
of
an
anthropomorphic
thing,
but
what
he
does
in
this
paper
is,
he
introduces
this
set
of
simulations
and
goes
over
a
number
of
days,
learning
in
lamarckian
and
then
darwinian
and
he
you
know
it
goes
for
500
Days,
on
learning
in
the
market,
because
they're
faster
processes,
but
goes
to
2500
days
in
the
darwinian
case.
And
then
he
has
this
camouflage
phenotype
which
actually
matches
an
example.
B
You
see
often
in
that
that
sort
of
demonstrates
natural
selection,
and
this
is
the
peppered
moth
example,
and
so,
if
you're
not
familiar
with
the
peppered
moth
example,
it's
that
you
have
these
moths
that
were
in
pre-industrial
England
and
they
had
different
Wing
colors.
And
then,
when
industrialization
happened,
there
was
a
lot
of
Colbert
and
a
lot
of
soot
on
the
on
the
branches
of
the
trees
and
the
darker
morphs
were
able
to
camouflage
better
with
the
soot
and
they
were
able
to
survive
at
a
greater
frequency.
A
B
And,
and
for
this
purpose,
I
think
it's
fine,
so
there's
this
camouflage
phenotype
that
he
uses
it's
lighter
or
darker,
and
the
idea
is,
is
that
if
you
run
the
simulation
for
so
many
days
using
these
types
of
transmission
and
feedback,
this
is
what
you
get.
So
you
end
up
in
a
in
the
darwinian
case.
You
end
up
with
these
straight
lines,
and
these
are
really
just
frequencies
of
the
phenotype,
so
lighter
lighter
the
lighter
phenotype
appears
early.
The
darker
phenotype
appears
early.
B
B
Now
that's
important
when
you
go
to
learning
in
the
markism
OR
lamarckian
Evolution,
you
see
something
really
interesting.
You
have
the
same
trend
line
that
is
basically
the
same
trend
from
median
phenotype
to
the
top.
You
have
these,
but
instead
of
having
these
frequency
bands,
you
have
this
sort
of
convergence
upon
this,
this
band
or
this
line
as
you
go
up.
So
in
learning,
it's
basically
learning
that
this
is
the
Phoenix
desired
phenotype
and
everything
goes
into
this
direction.
B
B
You
know
it's
producing,
it's
it's
moving
towards
that
Fitness,
but
it's
also
leaving
a
number
of
phenotypes
that
are
sort
of
not
so
tightly
selected
for
and
the
reason
that's
important
is
because
you
want
to
be
able
to
you
know
if
you're
evolving,
a
trait
you
don't
want
to
get
stuck
in
that
place.
You
want
to
have
enough
genetic
diversity
to
move
to
another
Fitness
imperative
later
on.
As
a
you
know,
you
have
stood
on
on,
you
know
on
the
branches
of
your
trees,
and
then
that
goes
away.
B
Then
you
don't
want
to
have
you
know
you
don't
want
to
go
to
extinction
because
you
can
no
longer
blend
into
the
tree
Bridge.
So
that's!
That's!
The
idea
here
and
so
learning
in
the
markism
OR
lamarckian
Evolution
will
sort
of
converge
upon
us
an
optimal
solution,
but
that
optimal
solution
comes
at
the
expense
of
this
ability
to
adapt
to
later
changes
in
the
environment.
A
B
A
B
A
B
B
Remodeling
melanism
is
a
polygenic
trait
and
so
the
mods
of
three
other
genes,
which
arranged
in
value
from
one
to
a
thousand
you
gave
the
genes
names,
Kudzu,
Harvard
and
Vanderbilt
Kudzu
is
a
growth.
Gene
and
Kudzu
is
linked
to
the
Harvard
Gene,
which
governs
the
rate
of
learning
and
the
Vanderbilt
Gene,
which
controls
an
inheritance
of
Acquisitions.
So
this
is
more
about
sort
of
these
modeling
sort
of
a
circuit
for
for
lamarchism
to
work,
and
so
this
gives
you
this
sort
of
feedback.
A
B
So
the
next
thing
I
want
to
talk
about
is
I've,
been
we
talked
about
this
several
weeks
ago,
and
this
is
a
paper
that
I
have
kind
of
prepared
for
this
special
issue
of
the
it's.
This
symmetries
and
symmetries
mines
of
matter,
I,
believe,
is
the
name
of
a
special
issue
and
I
got
an
email
from
the
person
who's
organizing
the
special
issue,
and
they
wanted
me
to
submit
something
from
our
group,
so
I'm
working
on
this
manuscript
and
it's
almost
ready
to
submit
as
a
first
draft.
B
It's
not
really
I
just
wanted
to
get
some
feedback
on
it.
So,
basically,
this
article
is
about
looking
at
development
and
symmetries
and
the
evolution
of
nervous
systems
or
the
development
of
nervous
systems
and
how
some
of
these
things
can
be
modeled.
So,
in
this
paper,
I
kind
of
go
through
some
of
this
I
go
through
actually
I
have
three
examples,
one
being
tensegrity
one
being
using
differentiation,
trees
and
the
third.
Using
these.
B
Let
me
see
these
these
embodied
hyper
graphs
that
I've
been
talking
about
this
past
summer.
So
this
is,
you
know.
These
are
three
ways
to
look
at
this
kind
of
type
of
system.
The
systems
defined
it's
a
developmental
system
that
generates
structure,
and
then
it
generates
these
nervous
systems,
these
nervous
system
networks
and
it's
all
defined
in
terms
of
cognition
and
some
of
the
things
we
can
learn
from
your
body,
cognition
specifically,
and
so
the
paper
kind
of
goes
through
building
that
model
and
then
bringing
in
these
three
models.
B
There
are
these
three
types
of
analysis
to
look
at
like
the
different
aspects
of
this
and
then
finally,
you
know
end
up
with
some
other,
like
sort
of
insights,
what
we
can
learn
from
it,
so
this
is
due
by
the
31st
I'll,
send
out
a
copy
of
it
for
people
to
look
at.
If
you
want
to,
you
know,
suggest
changes
or
things
like
that.
B
It's
pretty
much
close
to
the
word
limit.
It
was
like
8.
000
words,
I
think
is
the
limit.
The
target
Journal
is
the
Royal
Society
interface
Focus,
which
is
a
interdisciplinary,
Journal
and
they're,
going
to
have
a
number
of
different
papers
on
different
topics
related
to
symmetries
and
mind,
and
matter
and
I.
B
B
A
B
So
yeah
Susan's
been
sending
some
paper,
some
nice
papers
on
cellular
physics
and
cell
contractility,
and
so
I
figured
I'd
cover
some
of
these
papers
and
we
can
talk
about
them.
So
this
is
one
of
the
papers.
I
think
this
has
to
do
with
what
she
was
talking
about
with
having
a
heart
at
least
a
little
bit.
Maybe
the
Dynamics
of
microglial
polarization
reveal
the
resident
neuroinflammatory
responses
after
subarachnoid
hemorrhage.
B
B
B
B
Stuff
here
where
she
was
talking
about
yeah,
so
this
is
where
you
have
okay,
so
microglia
polarized
dynamically
from
M1
to
M2
phenotype.
So
they
have
these
two
different,
phenotypes
and
I.
Think
they're
related
to
just
their
basic
shape,
along
with
the
morphological
transformation
from
ramified
to
amoeboid
shapes
so
they're,
transforming
their
shape.
B
C
B
B
B
The
micro
wheel,
polarization
status,
is
distinct
in
molecular
inflammatory
responses.
M1
related
full
inflammation
was
predominant
in
the
early
phase
and
subsequently
transited
to
the
M2
related
anti-inflammation.
The
systemic
characterization
of
dynamics
of
microbial
polarization
contributes
to
the
understanding
of
the
origin
of
neuroinflammatory
responses
and.
A
B
B
B
It's
not
doing
its
job
after
it
takes
on
the
shape,
so
it's
like
this
flat
pancake
like
a
fried
egg
almost
and
so
it's
losing
its
rim
of
its
branches
that
go
off
like
this.
So
that's
interesting,
and
this
is
an
example
here
of
the
shape
from
day
one
to
day
ten.
So
in
day
one
you
have
this,
these
branches
that
go
out
towards
other
towards
neurons
and
other
parts
of
the
that
part
of
the
brain,
and
then
you
get.
B
This
almost
looks
like
these
irregular
branching
structures,
and
then
you
have
this
single
Branch
structure.
Here,
it's
not
really
branched
anymore,
that's
just
kind
of
like
a
big
mass,
and
then
you
get
these
almost
look
like
blebs
hermuboids
in
there,
where
they're
just
not
necessarily
doing
anything
they're
just
sitting
there
yeah.
So
that's
an
interesting
paper.
I
haven't
yeah
I
haven't
I'm
not
too
familiar
with
that
system,
so
I
I
I
can't
say
any
more
about
it.
B
B
B
B
The
biophysical
mechanisms
by
which
this
contractile
force,
or
these
contractile
forces
of
each
cell
cause
deformations
on
multicellular
length
scales,
is
not
fully
clear
so
to
investigate
this.
They
start
with
the
principle
Force
equilibrium
which
dictates
a
balance
of
tensile
forces
between
neighboring
cells.
Based
on
this
principle,
we
hypothesize
that
coordinated
changes
in
cell
area
result
from
tension
transmitted
across
the
cell
layer.
B
So
this
is
the
cell
area.
Well,
we
saw
an
example
in
the
last
paper
of
cell
shape.
Now
this
is
changes
in
cell
area
that
are
coordinated,
and
this
is
resulting
from
tension
transmitted
across
the
cell
layer.
So
I
imagine,
there's
a
tension
signal
or
some
signal
that
controls
tension
to
test
this
hypothesis.
Spatial
correlations
are
both
contractile
tension
and
the
Divergence
of
cell
velocities
were
measured
as
readouts
of
coordinated
contractility
and
Collective
area
changes
respectively.
B
So
this
is
coordinated,
contractility
and
Collective
area
changes.
That's
what
they're
measuring
experiments
were
designed
to
alter
the
spatial
correlation
of
contractile
tension
using
three
different
methods,
including
disrupting
cell
cell
adhesions,
modulating
the
alignment
of
actomyosin
stress
fibers
between
neighboring
cells
and
changing
the
size
of
the
cell
model
layer.
So
these
are
different.
B
Three
different
ways
that
they
did
this
cell
cell
adhesions
are,
of
course,
the
things
that
keep
cells
sort
of
in
contact
or
together,
and
that
can
introduce,
of
course,
stresses
and
tensions
across
different
cells
as
they're
pushing
against
one
another
and
sticking
to
one
another
modulating
the
alignment
of
active
myosin
stress
fibers
between
neighboring
cells.
So
the
stresses
are
transmitted,
I,
guess
across
the
cell,
the
different
cells
in
a
coordinated
manner
and
then
changing
the
size
of
the
cell
mono
air.
B
B
Bigger
or
smaller
in
all,
experiments
of
spatial
correlation
level,
tension,
Divergence
increased
or
decreased
together.
So
this
is
tension
and
Divergence
in
agreement
with
our
hypothesis,
I
guess.
Divergence
is
either
that
the
cells
are
coming
apart.
They're
less
they're,
adhering
to
each
other
less
to
relate
our
findings
to
the
intracellular
mechanism,
connecting
changes
in
cell
area
to
contractile
tension.
B
We
disrupted
activation
of
cellular
extracellular
signal,
regulated,
kinase
or
ERG,
which
is
known
to
mediate.
The
intercellular
relationship
between
cellular
and
contraction,
consistent
with
prior
knowledge,
a
temporal
cross
correlation
between
cell
area
and
tension,
which
is
over
time
of
course,
revealed
that
Earth
was
responsible
for
a
proportional
relationship
between
cell
area
and
contraction.
So
as
the
cell
area
increases
or
decreases,
contraction
increases
or
decreases
in
inhibition
of
Earth
activation
reduced
the
spatial
correlations
of
the
Divergence
of
cell
velocity,
but
not
of
tension.
B
So
this
is
where
work
is
inhibited.
The
correlations
spatial
correlations
between
Divergence
of
the
cell
velocity
is
maintained,
but
not
tension
so
together.
Our
findings
suggests
that
coordination
of
of
cell
contraction
and
expansion
requires
transfer
of
cell
tension
over
space
and
Earth
mediated
coordination
between
cell
air
and
contraction
and
time
so
this,
so
these
cell
contracts
need
to
be
coordinated
and
the
expansion
requires
transfer
of
cell
tension
in
space.
B
So
when
you
have
cells
in
a
in
a
you
know
a
cell
population
in
a
culture
dish
or
in
a
tissue,
those
sorts
of
things
need
to
be.
You
know:
cell
contraction
needs
to
be
coordinated
and
the
expansion
of
or
the
contraction
expansion
of
cells
requires
a
transfer
of
cell
tension
over
space,
so
tension
needs
to
be
coordinated
across
the
cells
and
then
irk
mediated
coordination,
which
involves
this
molecule
erk,
there's
a
coordination
between
celery
and
contraction
and
time.
So
this
controls
the
temporal
aspect
of
it.
B
So,
let's
see
if
we
can
find
some
so
yeah,
so
these
cell
culture,
for
this,
these
micro
patterning,
to
introduce
or
to
use
it
as
a
substrate.
So
micro
patterning,
is
where
they
put
a
pattern
on
the
on
the
culture
dish
and
they
allow
it
to
they.
They
have
like
conditions
where
it's
adhering
to
the
surface.
B
B
B
So
this
is
where
you
have:
the
cell
are
the
let's
see
this
color
map
is
kind
of
catching
my
eye
here.
So
this
is
where
you
have
okay,
so
the
first
one
is
root,
mean
squared
traction
and
control,
so
islands
and
Islands
treated
with
this
thing
that
they're
treating
it
with.
So
you
have
this.
This
is
in
pascals,
so
you
can
see
that
traction
is
increased
in
the
control
here
and
more
increase
in
the
control,
so
average
tension
versus
root
mean
squared
traction.
B
Tension
is
higher,
then
you
have
tension,
correlation
between
the
control
and
the
treatment,
the
treatment,
the
correlation
is
less
than
the
control
here,
and
you
can
see
that
the
correlation
length
is
higher
for
the
control
than
the
than
the
treated
actually.
This
correlation
actually
extends
over.
A
shorter
distance
scale
is
what's
happening
here,
so
it's
less
than
the
control,
but
it's
also
and
then
it
kind
of
catches
up
at
longer
length
scales.
So
I,
don't
know
if
there's
any
interesting
stuff
with
long
skill
going
on.
B
But
then
this
is.
This
is
of
course
E
and
F
are
where
you
have
these
different
islands
that
are
treated.
So
you
have
the
control
versus
the
treated,
and
you
can
see
it's
a
mosaic.
It's
not
exactly
in
one
spot
versus
another,
so
they're
in
different
patches
in
the
in
the
culture
and
then
G
and
H
are
these
spatial
correlation
of
of
the
div
Vector
of
control,
energy
ta
treated
Islands.
B
And
then
you
can
see
this
correlation
length
scale
difference
here
in
this
in
this
graph,
when
they
do
a
graphics,
they
usually
give
the
observations
of
these
dots
and
the
line
is
like
the
median
or
the
mean
just
so.
You
know
what's
going
on
there
in
this
case
here
this.
These
two
figures
here,
B
and
C-
are
confocal
images
of
stress,
Firearms,
fibers
of
low
serum
and
cno3
treated
conditions.
B
So
you
can
see
this
is
the
low
serum
condition.
This
is
the
treated
condition,
and
in
this
case
the
traction
is
much
higher
for
the
treated
condition.
The
average
tension
is
also
higher
you
get.
This
Croatian
of
the
treated
condition
is
higher
for
tension
at
low
length
scales,
but
then
falls
behind
the
control
at
higher
length
scales.
You
get
a
correlation
length,
that's
higher
in
cno3
and
then
in
Divergence
correlation
is
kind
of
higher
in
the
treated
up
to
150
microns,
and
then
it
falls
below
the
control.
B
And
then
you
get
this
difference
between
the
treatment
and
control
low,
serum
and
for
correlation
length.
So
it's
a
little
bit
different.
It's
actually
a
different
result
than
this
they're
using
two
different
things
here:
they're
using
egta
and
they're
using
cno3.
So
this
is
a
row
activator.
So
there's
a
lot
of
molecular
biology
here.
B
B
C
B
B
C
B
So
today,
we've
talked
about
groups
of
cells
where
you
have
physical
changes
that
change
their
shape,
change,
their
sort
of
structure
of
the
tissue,
that's
forming
changes
in
tissue
shape
and
those
are
all
interesting
things
and
I'm
going
to
change
gears
a
little
bit
and
talk
about
minimal
cells
and
how
they
their
structure
is
actually
useful
for
observing
some
of
these
Collective
processes,
and
this
paper
is
a
recently
published
paper-
nature,
Communications,
titled
in
vitro
assembly,
positioning
and
contraction
of
a
division
ring
in
minimal
cells.
So.
A
B
B
People
have
used
minimal
cells
in
different
ways.
You
know
it's
interesting
for
looking
at
sort
of
the
Baseline
mechanics
of
a
cell.
You
know
what
it
can
actually
produce
so
in
synthetic
biology,
they're
producing
a
lot
of
compounds
for
industrial
uses
and
so
and
and
also
looking
at
how
that
it
produces
things
like
proteins
and
said:
I,
don't
think
they've
used
minimal
cells
too
much
to
create
things
like
tissues
or
other
types
of
things
relevant
to
developmental
biology.
B
But
why
don't
we
read
through
this
paper
and
see
how
it
fits
into
what
we're
doing
so?
The
the
abstract
reads:
constructing
a
minimal
machinery
for
autonomous
salt
division
of
synthetic
cells,
there's
a
major
goal
of
bottom-up
synthetic
biology.
So
this
is
like
working
from
the
genes
upward
instead
of
working
say
from
the
tissues
down
to
the
genes,
we're
looking
at
how
the
tissues
and
the
phenotype
interact
with
the
genes.
B
B
They
want
to
assemble
a
contractile
ring,
so
they
want
to
assemble
something:
that's
going
to
contract
along
a
ring
and
they're
using
these
different
proteins
to
do
this
here
we
demonstrate
the
full
in
vitro
re
constitution
of
this
Machinery,
consisting
of
five
proteins
with
lipid
vesicles,
allowing
to
observe
the
following
sequence
of
events
in
real
time.
So
this
is
within
a
lipid
vesicle.
So
usually
these
minimal
cells
are
constructed
within
a
lipid
vesicle.
So
it
has
like
a
basically
a
membrane,
and
then
you
can
do
a
lot
of
different
things.
B
Some
people
have
experimented
with
kind
of
the
origins
of
Life
using
lipid
lipid
vesicles.
They
put
a
bunch
of
molecules
in
there.
They
add
an
electrical
charge,
they
they're
able
to
reconstruct
the
military
experiment
or
other
types
of
things
where
you
want
to
look
at
like
self-assembly.
So
these
are
very
useful
tools,
so
putting
these
five
proteins
within
a
lipid
vesicle
allows
you
to
observe
the
following
sequence:
events
in
real
time.
B
The
first
is
assembly
of
an
isotropic,
which
is
you
know,
having
different
properties
along
its
Network
filament
filamentous
ftsz
Network.
So
these
are
different
filaments
that
have
different
properties
in
the
network.
It's
condensation
and
or
bring
like
structure
number
two,
along
with
pull
to
pull
mode
selection
of
Min
oscillators,
resulting
in
Equatorial
positioning
and
three
on
set
of
ring
constriction,
deforming
the
vesicles
from
spherical
shape.
B
A
B
B
B
B
B
A
B
B
Ftse
is
your
own
to
polymerize
into
Dynamic
ring-like
structures
that
can
deform,
freestanding
membranes
and
co-reconstruction
of
the
mint
positioning
system
and
membrane
anchored
ftsz
on
supported
membranes,
which
can
form
the
spatial
regulation
of
ftsz
polymers
by
Min
patterns.
So
a
lot
of
this
is
spatially
registered
and
it's
being
constructed
in
this
ring
and
then
it
constricts,
and
then
you
get
this
division
event
and
moments,
so
they
talk
about
the
multi-protein
system.
B
So
this
is
your
your
system.
Here
you
have
the
lipid
membrane.
You
have
these
proteins
sitting
on
the
lipid
membrane,
so
this
membrane
is
at
the
edge
of
the
compartment
or
it's
at
the
bottom
of
the
compartment,
and
you
have
this
sort
of
these.
These
chemical
reactions
that
occur.
This
is
these
are
the
division
proteins
that
are
important.
B
These
are
purified
protein
cell
free
expression.
So
this
is
the
lipid
vesicle,
where
you
have
these
two
things
on
the
surface:
either
side
and
then
here's
the
ring
in
the
middle
that
forms.
So
you
get
this
ring
in
the
middle
between
these
two
side:
wall:
things
where
the
proteins
reside.
So
you
have
two
poles
on
the
lipid
vesicle.
You
have
this
ring
in
the
middle
and
this
is
driving
cell
division.
So
this
is
gonna.
These
are
almost
like
centromeres
and
an
actual
cell.
That's
what
they're
trying
to
replicate.
B
So
they
have
these
this
anisotropic
collection
of
these
fragments,
and
then
they
come
together
into
a
ring
here
and
then
the
ring
constricts
and
then
ostensibly
this
lipid
vesicle,
where
the
cell
would
divide
into
two
parts
according
to
these
poles.
So
you
can
see
that
this
ring
here
then
exists
and
lipid
vesicle
the
middle.
It's
acting
sort
of
like
a
centromere,
and
then
this
constricts,
these
poles
pull
apart
and
then
you
have
two
cells
and
you
have
the
protein
complex
on
either
side.
B
B
If
people
are
interested
in
getting
into
that,
and
so
they
look
at
different
crowding
conditions
on
Min
wave
Dynamics
inside
lipic,
lipid
vesicles
would,
you
importantly,
address
in
the
different
experimental
approaches
so
far,
so
they
try
different
things
to
look
at
that.
You
know
crowding
effect,
so
some
of
the
results
are
optimizing.
Ftse
and
Min
CDE
reconstruct
reconstitution
and
lipid
vesicles
under
macromolecular
crowding
conditions.
B
Another
Finding
is
a
co-reconstruction
of
Min
CD
and
ftsz
under
crowding
conditions
yield
the
Assembly
of
pronounced
FTS
Z
ring
structures,
so
they
have.
This
crowding
condition
is
not
required
for
fds
Z
ring
positioning,
so
you
find
that
there
are
some
proteins
that
aren't
directly
involved
are
required
for
certain
things,
but
they're.
Nevertheless,
in
that
Suite
of
minimal
in
that
minimal
cell
they're
included
in
the
minimal
cell.
But
we
don't
know
exactly
what
things
are
necessary
and
sufficient
for
different
processes.
B
B
A
ring
in
the
in
the
lipid
membrane
and
here's
some
more
images
of
this.
Where
you
see
the
expression
of
these,
this
has
been
CD
and
ftsz
being
co-expressed.
This
is
Mindy
and
ftsz
being
co-expressed,
so
you
get
to
sell
what
it
looks
like.
So
you
can
see
that
there
are.
These
proteins
are
being
expressed,
they're
going
to
form
this
ring.
This
ring
then
sits
at
the
middle
and
contracts.
B
B
B
B
Example
of
this
kind
of
oscillation
you're,
not
gonna.
This
is
not
gonna
result
in
a
differentiation
wave,
but
this
is
kind
of
interesting
how
these
oscillations
can
be
replicated
using
chemicals
of
Organization
principles.
So
this
is
a
pull
to
pull
oscillation
of
the
ring.
This
is
the
static
pattern
of
the
ring,
so
they
can
do
interesting
things
with
the
ring.
Aside
from
like
just
moving
to
the
middle,
you
can
see
things
that
are
you
know,
kind
of
you
know
it's
stable,
but
it's
not.
You
know
static.
Sometimes
it's
static.
Sometimes
it's
not.
B
B
These
chemical
self-organization
principles,
chemical
morphogenesis,
and
so
it's
a
very
interesting
how
they,
how
they
do
this
experimentally,
and
so
the
vesicle
diameter
is,
if,
if
you
bury
the
vesicle
diameter
and
make
it
bigger,
you
get
ring
structures
ring
deformation,
no
structures,
you
get
more
Rings,
plus
with
deformation.
As
the
vesicles
get
bigger.
You
get
ring
structure
more
ring
structures
period
as
the
vesicle
gets
bigger.
B
So
these
ftse
structures
increase
in
terms
of
their
frequency
as
you
get
a
bigger
vesicle,
so
it's
like
having
a
bigger
space
allows
for
more
things
to
happen,
and
then
you
have
this
aspect,
ratio
of
the
ring
and
then
the
diameter
before
deformation.
So
the
bigger
the
diameter,
the
smaller
the
aspect,
ratio
of
the
Ring
structure,
you
have
ring
just
plain
ring
structures
at
the
top,
with
a
high
aspect.
B
So
and
then,
if
you're
interested
the
references
at
the
bottom
so
I
think
that's
a
very
interesting
approach.
There's
some
if
you're
interested.
Actually
the
reference
section
has
some
good
references
on
synthetic
biology
and
some
of
the
state
of
the
art
in
that
area.
So
I
hope
you
learned
something
here
and
thank
you.
A
A
A
A
A
A
C
A
A
B
B
A
Right,
okay
or
you
can
collect
a
number
of
different
depths
things
like
that.
Okay,
they
typically
is
for
Dex,
but
light
field
is
the
concept
of
a
light
field.
If
you
look
around
any
room,
wherever
you
put
your
head,
you
can
pick
up
light
right,
okay,
so
a
light
Shield
is
all
of
the
light
photons
moving
around
the
room
at
once,
all
the
different
angles.
A
B
Just
a
few
words
on
light
field
Imaging.
So
if
you're
interested
in
like
field
Imaging,
we
have
a
few
more
references
for
you.
The
first
one
is
the
Wikipedia
page
and
light
field
camera,
so
light
field,
camera
or
a
planoptic
camera.
There's
a
camera
that
captures
an
information
about
the
light
field
emanating
from
a
scene.
So
it's
basically
picking
up
the
light
coming
from
a
scene
at
various
wavelengths.
So
it's
collecting
all
that
information.
Then
inside
the
camera
it's
taking
different
parts
of
that
information
and
processing
it.
B
So,
for
example,
you
can
we'll
see
in
a
bit
you
can
change
the
focus.
You
can
do
other
things
you
can
take
the
information
you
want
out
of
that
image.
So
one
type
of
one
type
of
light
field
camera
uses
an
array
of
micro
lenses
placed
in
front
of
another
as
conventional
image,
sensor
to
sense,
intensity,
color
and
directional
information,
multi-camera,
Rays
or
another
type,
and
then
even
Holograms,
which
we're
familiar
with
are
a
type
of
film-based
light
Field
image.
B
So
hologram
is
where
you
have
you
record
a
wavefront,
and
then
you
reconstruct
it
in
pieces.
So
you
can
generate
three-dimensional
images.
So
this
is
an
example
here
of
light
field
photography.
This
is
a
dog
with
a
ball
on
the
floor
and
you've
taken
this
light,
Field
image,
and
you
can
change
the
focus
of
the
image,
because
you
have
all
that
information
collected.
You
just
need
to
process
it.
Foreign.
B
B
Then,
maybe
about
10
years
ago
it
came
on
the
market
for
the
first
time
and
it's
been
with
us
ever
since
I
had
a
remembered,
an
article
from
the
IEEE
spectrum
that
I
remember
reading,
but
it
turns
out
it
was
like
a
decade
ago
now
so
I
wanted
to
find
a
newer
reference
and
I'll
talk
about
that
in
a
minute.
You
also
have
your
focused
planoptic
cameras,
so
you
can
position
your
lenses
in
different
places
and
you
can
play
around
with
it.
B
You
can
get
these
kinds
of
image
effects
and
you
can
then
collect
data
that
have
like
things
are
different
full
in
different
focal
planes.
So
you
can
see
how
this
isn't
useful
for
biological
Imaging
as
well.
You
can
do
you
can
capture
different
aspects
of
the
thing
that
you're
Imaging
a
different
focal
lengths
and
different
different
properties
of
the
image.
B
We
actually
have
that
now
with
fluorescence
Imaging,
but
this
is
of
course,
they're
a
little
bit
different
animal
because
you
can
actually
put
these
things
together
as
part
of
the
main
image
instead
of
stitching
it
together,
post
hoc.
So
there
are
a
number
of
prototypes
that
have
been
created.
There
are
all
sorts
of
people
like
creating
prototypes
of
innovative
ways
to
do
this
type
of
Photography.
This
is
very
useful.
This
is
also
a
computer
Graphics
area
of
computer
Graphics
research.
B
So
this
is
the
article
from
the
IEEE
Spectrum
from
2020,
and
this
one
actually
goes
into
an
immersive
display,
creating
panoramic
virtual
screens.
So
this
is
a
not
not
a
primer
on
light
field.
Imaging
per
save.
This
is
a
immersive
display
that
uses
light
field.
Imaging
as
its
source,
so
just
imagine,
wearing
a
200
gram
object
on
your
face
for
6.5
hours.
B
It's
really
exhausting,
but
6.5
hours
is
the
average
time
we
spend
in
front
of
computers
easily
every
day,
and
so
this
this
led
this
inventor
to
come
up
with
a
some
level
of
immersion
in
a
display
that
you
don't
have
to
wear
on
your
face.
So
this
is
like
merging
virtual
reality
and
this
sort
of
light
field
Imaging,
because
the
information
that's
captured
through
light
field
Imaging
can
be
processed
in
different
ways
to
make
it
immersive,
because
you
have
many
many
different
fields
of
view
and
you
can
render
all
of
those.
B
So
this
is
the
this
goes
through
this
technology,
this
this
application
here
and
so,
whereas
conventional
splays,
direct
flat
images
at
its
viewers,
it's
light
field
display,
creates
a
Windows
like
3D
scene
by
recreating
the
field
of
light
rays
that
might
travel
from
every
point
in
every
direction
within
3D
space.
So
in
a
3D
in
a
White
Field
camera
you're
capturing
things
from
a
three-dimensional
space
and
every
ray
of
light.
That's
coming
their
angles,
so
you
have
angle
sensitive
pixels.
B
You
have
things
that
are
straight
ahead
and
then
you
can
render
it
in
the
same
way.
So
so-called
Auto
stereoscopic
displays
these
pixels.
That's
Enlighten,
specific
directions
to
project
a
different
image
that
each
viewer
add
different
image
each
of
a
viewer's
eyes.
So
you
basically
have
two.
You
know
you
have
two
eyes:
you're,
projecting
things
into
those
two
eyes
and
if
you've
ever
were
to
Via
our
headset,
you
know
that
there
are
two
cameras
in
there:
the
two
lenses
that
you
look
through
and
because
you
have
stereoscopic
Vision,
you
merge
those
images.
B
B
Stereoscope
at
the
multifocal
displays
for
3D
applications
in
in
this
application,
though,
they
want
to
create
a
virtual
panoramic,
2D
image
and
to
deliver
it
to
the
viewer.
They
call
it
a
near
head
display.
So
it's
not
a
heads
up
display,
which
is
what
we
usually
use
in
VR,
but
near
head
display,
and
so
you
can
curve
the
display
to
take
advantage
of
what
you've
captured
in
your
camera
and
project
it
to
the
person
looking
at
it.