►
From YouTube: DevoWorm #30: Project Updates, Diatom Diversity and Modes of Movement, Push-Pull Morphogenesis III
Description
Discussion on Diatom cell tracking, physical movement mechanisms, and interspecific diversity. Push-pull Morphogenesis III: bioprinting optimization and a systems biology for development. GsoC project updates, Attendees: Alon Samuel, Richard Gordon, Bradly Alicea, Susan Crawford-Young, Morgan Hough, Harikrishna Pillai, Wataru Kawakami, and Sushmanth Mereddy
C
B
All
right
so
yeah
welcome
to
the
meeting
alan
and
dick
and
susan
and
otaru
and
morgan
so
yeah
today,
we'll
talk
about
maybe
some
g-shock
updates,
there's
hare
krishna
so
and
then
we'll
get
into
some
other
things.
I
know
we've
had
an
email
conversation
about
diatoms
and
working
on
diatoms,
so
maybe
we'll
go
over
that
a
little
bit
and
then
some
updates
and
then
we'll
maybe
do
some
papers.
B
A
A
C
A
E
D
E
A
E
C
E
F
E
So
I
downloaded
the
digi
the
repo
I
kind
of
installed
it
locally,
and
I
saw
all
the
kind
of
like
prerequisites.
It
kind
of
took
a
bit
of
time.
And
then
I
trained
the
detectron
segmentation
and
object
detection,
the
neural
network,
and
that
gave
some,
I
think,
not
like
results.
That
kind
of
like
I'm
not
like
happy
with,
because
for
like
tracking.
E
If
it's
smooth,
I'm
kind
of
like
wanting
something
to
kind
of
like
get
all
of
it,
garlic,
maybe
in
a
high
precision,
which
is
a
bit
hard,
in
instance,
segmentation
to
kind
of
differentiate
between
the
different
instances
of
the
diatoms
and
yeah
we're
kind
of
like.
I
think,
after
that,
we
chatted
a
bit
in
the
email,
and
we
said
that
we're
going
to
do
a
meeting.
A
C
A
The
based
on
its
outline
and
we
were
able
to
get
subpixel
resolution
doing
that.
Okay,
because
as
long
as
the
diatom
doesn't
turn
or
roll
over
in
some
sense
you,
then
you
have
the
same
view
and
therefore
you
can
track
it
easily
yeah
relatively
easily.
But
so
we
we
did
an
analysis
now.
We
also
took
some
plastic
beads
and
centered
them
to
a
glass
surface.
I
think
so
we
would
get
an
image
of
something
that
was
not
moving.
E
A
E
E
A
A
E
A
A
E
E
Exactly
because
like
he
did,
I
think
he
did
something,
and
then
he
got
like
some
noise
in
the
in
the
velocity
of
the
y
and
the
y
axis,
or
something
but
yeah,
just
just
to
see
like
what.
What
and
I'm
kind
of
like.
I
don't
know
like
if
I
think
we're
gonna
have
to
decide
of
a
definition
of
kind
of
like
what
smooth
is
or
like
if
it's
gonna
move
like
what
we
got
like
for
now
or
like
we
can
say,
like
that's
the
smoothest
that
we
can
so
like.
Okay,.
A
E
A
E
A
C
E
C
E
E
C
C
B
Yeah,
it
sounds
great.
Thank
you
for
that
update.
We
had
such
month
put
something
in
the
chat
here.
Actually,
rutherford
did
this
thing
in
gold,
foil
experiment.
He
was
in
dark
room
for
more
than
one
week
to
do
gold
foil.
So
this
is
the
gold
foil
actually
sushma.
If
you
want
to
introduce
yourself
and
tell
us
more
about
that,
could
you
do
that?
G
Junior
of
my
death,
like
he
was
my
senior
he
introduced
me
to
devolent
organization
and
all
I
found
it
so
fascinating.
Actually,
when
dick
was
talking
about
being
in
a
dark
room,
allen
was
telling
like
that
thing.
Only
rutherford
did
in
gun,
foil
experiment
to
see
those
photons
like
they
are
moving
or
not.
G
G
B
G
G
I
was
able
to
start
working
on
that
big
issue
which
was
issued
by
my.
I
was
laughing
the
architecture
of
full
complete
devil
and
I
need
to
meet
the
my
today
for
explaining
about
all
stuff
how
it
works,
but
I
couldn't
go.
He
was
in
lighting
so
today,
tomorrow,
maybe
I'll
meet
and
I'll
start
working
on
that
issue.
G
G
G
B
A
A
We
don't
know
if
the
elastic
properties
of
that
or
the
viscoelastic
properties
of
it
might
influence
the
cells
and
therefore
provide
a
feedback
mechanism
between
the
cells,
because
right
now
there
is
no
obvious
way
that
the
cells
can
communicate
with
one
another.
But
nevertheless
the
emotion
is
nearly
synthetic.
A
There's
speculation:
they
do
so
by
light
piping
could
be
mechanical
through
this
jelly.
That's
on
the
outside.
You
don't
know,
okay,
so
an
attempt
to
visualize.
I
don't
know
if
you
can
see
it.
I
I
can
just
see
some
particles
moving
suggesting
that
they're
part
of
the
jelly
mask,
but
enhancing
it,
maybe
with
histogram
equalization
or
something
like
that
might
show
it
up.
E
C
A
Yeah
bradley
there
was
also
an
interesting
movie
that
came
across.
I
forgot,
which
one.
But
you
know
the
synchronization
is
only
partial
and
there's
actually
an
image
where
two
parts
of
the
same
column
collide
into
each
other
and
stop
suggesting
that
there
is
a
means
of
some
sort
for
long-range
communication
between
the
cells,
not
just
near
as
neighbors.
C
E
B
A
A
E
E
E
A
A
C
A
A
B
H
H
Which
I
created
using
the
python
code,
so
the
next
step
was
to
map
the
image,
so
I
I
wasn't
able
to
initially
I
wasn't
able
to
do
using
the
code
itself
because
the
results
were
not
nice.
So
what
I
did
was
I
loaded
the
model
into
blender
and
then
this
was
the
best
uv
underwrap,
which
I
could
get,
which
perfectly
fits
over
the
image
almost
perfectly
so
so
now.
H
So
if
you
arrange
it
like
this,
the
image
will
be
mapped
correctly.
So
this
is
what
I
worked
on
and-
and
I'm
also,
I
also
worked
on
parts
of
application
of
my
application,
which,
which
is
the
final
result
so
yeah
and
now
currently,
what
I'm
doing
is
finishing
up
the
application
and
also
writing
the
readme
file
for
the
github
repository
okay.
B
H
H
B
H
H
B
B
H
C
A
The
vessel
functions
are
over
a
circular
plane
now
on
a
sphere
we're
dealing
with
spherical
harmonics.
So
my
question
is
given
that
these
are
projection
images
of
a
spherical
surface.
Is
there
some
relationship
between
vessel
functions
in
the
plane,
with
spherical
functions,
spherical
harmonics
on
the
sphere
which
we
could
exploit
in
order
to
project
the
image.
A
H
A
D
D
B
B
Koran
as
well
and
make
sure
that
you're
coordinated
for
the
you
know
the
final
submission
and
everything
well
I'll
talk
about
that
more.
You
know
in
the
slack
and
we'll
we'll
sort
of
coordinate
that
make
sure
that
everything
is
okay,
so
yeah,
it's
good,
okay
and
there's
a
link
but
morgan.
I
think
that's
that
book
he
just
mentioned.
C
C
B
C
B
Okay,
well
I
mean
you
know
you
can
do
like.
I
guess
it
just
has
to
run
and
it
doesn't
need
to
be
like
perfect.
So
I
wouldn't
worry
too
much
about
testing
just
make
sure
that
it
can
run.
You
know
at
least
just
get
it
yeah,
and
then
that's
that's
fine
for
now
and
then
after
the
submission
deadline,
then
you
can
work
on
it
more.
If
you'd,
like
you
know,
that's
that's
always
the
way
these
things
go.
B
They
give
you
a
certain
amount
of
time
to
to
get
to
a
certain
point
and
then
that's
it,
but
we
have
the
deadline
coming
up
on
the
seventh
once
again.
So
do
you
think
that
you'll
be
ready
to
submit
by
then
and
if
both
of
giahong
and
yourself
have
to
submit
separate?
You
know
reports,
you
have
to
submit
something
and
you
have
to
have
the
documentation
and
everything.
So
do
you
think
you're
on
track
for
that.
C
B
B
B
Maybe
before,
if
you
want
and
then
you
can
submit
it
and
then
you
know
you
can
put
the
what
you're
you
and
your
hunger
been
doing
into
a
larger
package,
or
I
guess
it's
all
one
sort
of
all
one
package.
Now
it's
just
a
matter
of
what
part
you
did
so
like
you
focused
on
stage
one
and
jiahung
focused
on
stages,
two
and
and
three.
I
guess
you
kind
of
worked
on
together,
but
you
know
that's.
B
C
B
All
right
so
and
then
I
wanted
to
mention
one
more
thing
about
that
about
the
the
graph
neural
network
stuff.
So
I
I've
forgotten
about
this
a
little
bit
over
the
summer,
but
I
think
we
can
still
do
it
so
there's
this
conference
that
gia
hang
brought
to
my
attention-
and
this
was
this
conference
on
learning
on
graphs.
B
B
B
So
it's
kind
of
an
odd
system,
but
you,
I
think
basically
they
want
to
make
sure
that
you're
submitting
an
abstract
that
isn't
like
off
the
wall,
something
they
obviously
wouldn't
want.
And
then
you
can
submit
your
paper
and
then
you
know
the
conference
is
on
december
9th.
So
this
is
like
you
know,
we'll
get
a
final
decision
by
like
sometime
in
november.
B
Actually
I
think
like
a
submission
decision
in
october
and
then
some
revisions
in
november
and
then
there's
a
final
decision
and
then
the
conference
is
december
9th.
So
this
would
be
nice
to
have
like
as
a
you
know,
something
that
just
is
an
immediate
output
of
this
of
this
summer
project.
So
these
accepted
proceedings
papers.
I
don't
know
what
the
acceptance
rate
is
will
be
published
in
proceedings
for
machine
learning
research.
B
So
that's
that's
nice!
Full
proceedings
papers
can
have
to
nine
have
up
to
nine
pages
for
with
unlimited
for
reference
as
an
appendix.
So
again,
like
you
know,
all
the
work
that
jihang
and
and
whatara
have
been
doing
that.
That's
probably
gonna,
be
quite
a
bit.
You
know
it'll,
probably
not
be
too
hard
to
get
near
nine
pages
on
that,
along
with
some
some,
you
know,
information
about.
Why
we're
doing
it
so
the
I
think
the
registration
is
free
and
it's
on
in
a
virtual
conference.
B
So
it's
very
little
overhead
on
our
part.
You
know
just
sign
up,
I
guess
and
then
there's
an
overleaf
template
which
we
can
use
as
well.
But
you
know
that's
those
are
the
details.
We'd
work
out
later
so
yeah
this
is
and
then
I
looked
at
the
subject
areas
and
it
looks
like
we
kind
of
fit
in
here
somewhere.
B
All
right
there
you
go
and
that's
a
I
think,
an
immediate
goal
for
this
and
then
longer
term.
You
know
well
longer
term
being
like
you
know
soon
after
we'll
try
to
integrate
this
into
the
diva
learning
platform
and
see
how
that
goes,
and
then
then,
there's
also
a
diva
learn
sort
of
technical
paper,
we'll
probably
update
from
there
too,
because
if
we
update
the
software
we'll
update
that
so
yeah,
that's
that's
the
conference
call
for
papers.
B
So
that's
that's
about
that
now
I
wanted
to.
I
know
that
hari
krishna
had
to
leave.
Thank
you
hari
krishna,
and
thank
you
sushma
for
attending.
I
also
had
some
things
on
diatoms.
I
I
pulled
up
because
I
knew
we'd
be
talking
about
them.
This
might
be
a
benefit
to
alan,
because
I
don't
know
how
much
he
knows,
but
he's
seen
of
diatom
phenotypes
and
things
like
that.
So
there's
a
wikipedia
page
on
diatoms.
B
B
They
actually
have
these
different
types
of
morphology,
the
pennate
one,
which
is
long
in
the
centric
one.
So
the
basilaria
are
more
penny
and
then
this
is
the
structure
of
a
fristal
which
is
the
where
they
have
these
micropores,
and
they
have
all
this
detail
that
dick
was
talking
about
when
he
said
that
the
details
are
fine,
you
need
a
huge
you
would
have
to
if
you
were
to
do
a
3d
printing
job
of
like
the
deep,
the
anatomical
detail
would
be
like
50
meters
in
size,
because
all.
B
B
Oh
wow,
okay,
so
yeah,
and
this
is
a
shape
classification,
so
they've,
you
know
they're.
The
rod
shaped
ones
like
this.
This
is
kind
of
like
basil
area,
but
they
have
the
pores
along
the
side
and
some
other
features.
Then
they
have
the
flake
shapes
which
are
like
this.
You
see
they're
really
diverse
in
terms
of
their
shape
and
then
the
three-dimensional
shape.
So
some
of
them
have
like
depth.
B
And
so
then
their
behavior,
which
we
were
talking
about-
and
this
is
just
the
the
wikipedia
page.
So
this
isn't
like
you
know,
complete
by
any
means,
but
they
talk
about
how
in
in
I
know
dick
has
written
some
work
on
like
the
different
modes
of
movement.
So
they're,
one
of
the
things
about
diatoms,
is
that
there
are
a
lot
of
theories
about
how
they
move
and
there
are
a
lot
of
potential
mechanisms,
but
no
one
actually
knows,
and
it
varies
by
species.
B
I
imagine,
but
no
one
actually
knows
what
the
definitive
mechanisms
are
like
that's.
Why
that's?
Why
we're
kind
of
like
why
we're
doing
this?
Because
there's
a
you
know,
we
have
the
different
modes
of
movement
and
we
don't
know.
I
mean
we're
not
going
to
get
into
like
the
mechanism
through
just
observing
videos,
but
you
know
just
characterizing.
The
movements
are
enough
and
then
knowing
exactly
what
the
mechanism
is.
That's
that's
actually
quite
hard,
so
yeah.
B
B
B
So
this
is
this
sort
of
drawing
under
you
know,
looking
under
a
microscope
and
drawing
out
the
detail,
and
then
you
know
there,
people
have
done
this
in
various
ways,
just
observing
these
different
cell
types
and
that
you
know
that
can
give
you
some
information
too,
as
to
what's
going
on.
They
have
a
good.
They
have
a
interesting
worldwide
distribution.
C
B
Yeah,
so
this
is
a
scientific
american
article
on
diatom
or
the
trouble
with
life
in
glass
houses,
and
so
a
lot
of
this
is
this
kind
of
goes
over.
This
diversity
of
diatoms
they're
made
of
the
silicates
that
they're
kind
of
glad.
You
know
they
have
these
glass
houses,
and
this
is
an
art
nice
article
on
some
of
the
you
know.
What's
going
on
with
this,
it's
a
scientific
american
level.
B
B
G
D
Yeah,
nice
and
so
I've
got.
I've
got
some
pictures
from
my
phone
that
I'll
I
can.
I
can
get
out
as
well
as
like
they're.
You
know
pictures
of
their
microscopy
set
up
but
like
because
they
are
a
a
community
resource.
Right
was
trying
to
be
an
educational
like
I'm
gonna
make
a
time
to
set
a
time
to
talk
with
their
with
their
researchers,
and
you
know
maybe
get
them
to
to
come
talk.
You
know.
C
B
B
B
So
in
the
past,
I
think
it's
been
several
meetings
now,
since
we
got
into
this.
We
had
this
thing
about
push
and
pull
morphogenesis,
so
I
think
it's
in
the
recordings,
if
you
don't
remember
it,
but
is
this
a
series
of
papers
where
they
talk
about
this
mechanism
for
morphogenesis,
where
there's
like
bending
of
the
tissue
and
then
there's
a
feedback
that
triggers
gene
expression
which
triggers
more
bending
and
then
the
definition
of
different
things.
B
This
is
different.
This
is
where
you
have
these
spots,
where
there's
buckling
and
then
the
buckling
is
reinforced
through
different
mechanisms,
and
then
you
get
the
stippling
pattern,
so
it's
very
different.
Then
there,
I
think
you
know
we
did
a
couple
papers
on
that
and
then
now
there's
this
other
paper.
I
wanted
to
talk
about
within
this
sort
of
the
in
the
scope
of
this,
and
this
paper
is
from
the
international
society
of
biofabrication
or
I
guess
it's
bio
fabrication
journal.
B
B
So
the
abstract
reads:
during
bio:
printing
cells
are
suspended
in
a
viscous,
bio
ink
and
extruded
under
pressure,
there's
small
diameter
printing
needles,
the
combination
of
high
pressure
and
small
needle
diameters
exposed
cells
to
considerable
shear
stress,
which
can
lead
to
cell
damage
and
death
approaches
to
monitor
and
control.
Shear
stress
induced
cell
damage
are
currently
not
well
established
to
visualize
the
effects
of
printing
induce
shear
stress
on
plasma
membrane
integrity.
B
We
recommend
that
bioink
should
be
routinely
supplemented
with
physiological
concentration
of
calcium
ions
to
reduce
shear
stress
into
cell
damage
and
death
during
extrusion
bioprinting.
So
this
is
kind
of
like
a
nice
paper
on
talking
about
some
of
the
techniques
they
use
for
bioprinting
and
some
of
these
physical
aspects
to
it.
B
So
they
talk
about
fluid
shear
stress.
They
talk
about
how
to
make
the
cells
more
robust
to
this.
So
basically,
what's
happening,
is
they
want
to
print
cells
on
a
substrate
and
they
force
it
through
a
needle
or
some
sort
of
deposition
device,
and
one
of
the
problems
they
have?
Is
that
the
cells
break
apart
and
then
it's
no
longer
any
good.
So
what
they
want
to
do
here
is
they
want
to
you?
You
utilize
different
mecha.
B
You
know
different
types
of
things
to
mitigate
these
cells
blowing
up
when
they
shoot
them
out
so
they're
putting
them
in
in
high
pressure
situations,
so
they're
sending
them
down
this
printing
needle
and
they're,
depositing
them,
and
you
have
this
sort
of
they're
doing
these
things
with
with
the
plot
of
the
chemicals
and
and
different
things
that
they're
doing
on
this.
So
I
don't
know
if
we
have
any
good
pictures
of
this
process,
but
it's
you
know
it
has
a
lot
to
do
with
microfluidics
and
some
of
these
other
things.
B
F
B
I'm
not
yeah,
I'm
not
not
really
sure
I'm
not
familiar
with.
I
just
never
familiar
with
this
technique
so,
but
that
yeah,
I
guess
it
would
make
some
sort
of
difference.
I
mean
but
anyways
yeah.
So
this
is
this
just
kind
of
goes
through
this
and
I
can
send
this
along
if
you're
interested
in
reading
more
about
it
yeah
so
next
paper.
I
want
to
talk
about.
B
B
But
this
is
a
review
article
on
pattern,
growth
and
control
and
some
of
the
things
that
are
involved
in
pattern
formation.
So
this
is
by
arthur
d,
lander,
uc,
irvine
and
the
abstract.
Read
systems
biology,
seeks
not
only
to
discover
the
machinery
of
life
but
to
understand
how
such
machinery
is
used
for
control
and
when
they
say
control,
they
mean
regulation
that
achieves
or
maintains
a
desired
useful
end.
B
So
it's
like
you
know
when
you
have
some
sort
of
regulation
in
the
cell,
whether
it
be
a
forces
or
of
gene
products
or
of
like
you
know
some
sort
of
like
tissue
or
something
there's.
This
idea
of
control,
there's
regulation,
which
is
making
sure
that
it,
you
know,
is
sort
of
in
this
range
of
functional
range
and
then
there's
control,
which
is
this
sort
of
desired.
Useful
end
aspect.
B
So
it's
like
control
is
always
making
sure
that
it's
at
a
that,
what
the
output
is
useful
instead
of
just
regulating
it
like
you,
know,
cutting
off
some
process
and
then
it
doesn't
really
have
a
useful
end
to
it.
So
this
is
actually-
and
it
has
an
engineering
centered
approach.
So
this
sort
of
goal,
directed
engineering-centered
approach,
also
has
deep
historical
roots
in
developmental
biology.
So
this
is
something
we
see
in
engineering
and
we
can
see
in
bioengineering,
of
course,
but
you
you
see
this
metaphor
being
used
in
developmental
biology
in
a
way.
B
You
keep
the
room
temperature
at
73
degrees,
fahrenheit,
you,
the
room,
temperature
fluctuates
and
then
it's
up
to
your
climate
control
systems
to
bring
it
back
down
to
that
temperature
back
up
to
that
temperature.
So
that's
that's
the
kind
of
thing
that
they're
looking
at
here.
So
one
of
the
things
anyways,
the
usefulness
of
self-organizing,
behavior
and
we've
talked
about
self-organizing
behavior,
with
respect
to
like
a
different
swarm,
and
you
know
swarms
and
flocks
and
other
types
of
collective
intelligence
and
other
types
of
collective
behaviors.
B
B
It
could
be
just
something
like
we
talked
about
earlier
that
if
something
moving
in
a
column
of
water
just
kind
of
at
random,
that's
noise,
you
know,
there's
also
technical
noise
like
static
or
something
like
that,
but
in
in
in
biology,
noise
is
actually
very
important,
and
sometimes
it's
it's
driven
intentionally
to
achieve
certain
things
in
the
in
the
system.
So
sometimes
there's
a
sort
of
intrinsic
noise.
That's
driven
by
stochastic.
B
And
then,
finally,
the
absence
of
a
free
lunch,
so
free
lunch
is
actually
where
you
get
basically
work
for
free
or
you
get
energy
for
free,
and
so
that's
something
that
if
you
know
anything
about
like
in
computer
science,
there's
the
no
free
lunch,
theorem
or
free
lunches
are
usually
like.
You
know,
energy
or
work
or
information
for
free
and
that's
usually
a
violation
of
our
principles
like
the
laws
of
thermodynamics
and
things
like
that.
B
So
people
are
very
suspicious
when
you
say
you
know,
when
you
mention
something,
it
sounds
like
a
free
lunch,
not
just
because
they
don't
like
to
give
out
free
lunches,
but
because
it
does
violate
some
of
the
principles
of
our.
You
know
our
our
theories
about
you
know
thermodynamics
and
information.
B
By
illuminating
such
principles,
systems
biology
is
helping
to
create
a
functional
framework
within
which
to
make
sense
of
the
mechanistic
complexity
of
organism
on
development.
So
this
is
kind
of
like
a
nice
review
of
going
through
some
of
these
print.
Some
of
these
ideas
that
are
being
maybe
imported
from
systems
biology
and
you
know
bringing
them
into
developmental
biology.
B
The
mainstream
of
developmental
biology
is
not,
typically,
they
don't
think
like
systems,
biologists,
necessarily
their
focus,
isn't
on
some
of
these
larger
concepts
of
light
control
or
of
like
you
know,
noise
or
robustness,
or
any
of
that.
So
this
is
kind
of
a
you
know,
a
newish
area
in
that
field.
Now
some
you
know
in
like,
for
example,
in
development
or
in
evolutionary
biology.
B
You
know
there
are
people
working
on
things
like
robustness
and
evolvability,
but
even
those
aren't
necessarily
in
the
mainstream
of
that
field.
So
sometimes
you
get
these
ideas
that
kind
of
enter
the
field
in
different
ways
and,
and
you
don't
necessarily
know
their
source.
So
as
a
consequence,
people
tend
to
misuse
terms
where
they
tend
to
misapply
the
concept.
B
So
this
is
like
a
nice
way
to
kind
of
ground
people
and
where
these
things
come
from
yeah,
so
the
frequency
one
one
part
here
mentions
the
frequency
and
degree
with
which
embryonic
regulation,
canonization
robustness
and
precision
are
encountered
and
development
raises
many
questions.
So
this
is
one
of
the
things.
That's
fascinated
people
about
like
say
embryogenesis
is
that
you
can
get.
B
You
know
these
precise
things
going
on
in
development
and
they're
self-organized
in
the
sense
of
like
how
we
might
see
like
pattern,
formation
being
self-organized
and
people
ask
questions,
and
you
know
the
the
question
is
you
know?
Is
it
something
that's?
What
are
the
mechanisms
here
at
play?
They're,
not
things
that
you
can
necessarily
probe
from
like
the
typical
developmental
biology
experiments
or
there's
people
done
these
experiments
to
show
you
know
proof
of
concept
things,
but
some
of
these
things
like
robustness,
it's
kind
of
hard
to
measure
that
from
a
typical
experiment.
B
So
you
know
you
have
to
kind
of
consider.
Maybe
what's
going
on
underneath
the
hood.
So
are
there
common
principles
underlying
all
such
phenomena?
Are
they
conserved
mechanisms?
Can
we
explain
how
and
why
such
processes
evolved?
So
you
know
all
these
questions
revolve
around
this.
These
processes
and
you
could
say
well,
of
course
they
do.
You
don't
have
to
waste
your
time,
we
kind
of
know
that
they
do.
B
But
you
know
you
have
all
these
different
species
and
I've
shown
you
like
these
different
species
just
of
diatom
and
all
these
different
phenotypes
and
the
question
is,
is
you
know,
is
this
a
common
principle
underlying
all
these
different
phenotypes?
All
the
different
phenotypes
that
we
see
just
in
in
diatoms
are
the
same
or
the
same
processes
underlying
some
of
these.
You
know
changes
in
morphology
or
changes
in
behavior,
and
you
know
that's
still.
I
guess
we
assume
that
they
are,
but
that's
not
necessarily.
That
may
not
necessarily
be
the
case.
B
There
may
be
mechanisms
that
are
unique
for
certain
species.
There
may
be
some
that
are
conserved
or
shared
among
species
in
other
cases,
so
this
is
something
to
think
about.
So
a
lot
of
what
we
think
about
in
terms
of
these
universal.
Maybe
we
can
call
them
universal
principles
that
we
mentioned
up
in
the
abstract.
B
You
know
we
think
about
this
as
sort
of
a
like
a
physical
law,
but-
and
there
are
physical
laws
that
sort
of
govern
some
of
the
things
in
development,
so
like
fix
a
lot
of
diffusion
or
the
universal
gas
law,
and
those
are
things
that
will
give
you
some
of
these
products
of
emergence
so
yeah.
We
have
a
lot
of
things
that
exist.
We
can
apply
them
to
biological
systems,
but-
and
this
is
something
I've
asked
myself-
you
know-
maybe
these
aren't
really
sufficient
for
really
explaining.
B
What's
going
on
the
biology,
maybe
they're,
maybe
they're
not
like
totally
sufficient.
Maybe
they
can
actually
give
you
a
nice
heuristic,
but
you
know
maybe
we
need
special
sort
of
corollaries
of
physical
laws
or
biological
systems.
It's
not
really
clear.
So
they
have
these.
This
nice
diagram
of
control
objectives
in
morphogenesis.
B
You
start
from,
let's
see
so
this
is
going.
This
talks
about
feedback
regulation
and
some
of
these
control
objectives,
so
you
can
see
across
the
sort
of
these
different.
You
know
this
is
an
these
are
embryos
here.
These
are,
you
know
they
start
to
get
wings.
Drosophila
wings,
you
get
phenotypes,
you
get
brains,
you
get
muscle,
striated
muscle,
you
get
a
liver
here
and
you
get
like
a
retina.
B
So
you
have
these
different
things
in
the
in
that
are
developing
either
at
the
scale
just
kind
of
like
this
embryo
or
these
organs
or
organisms.
And
so
you
have
this
circle
here,
where
it
tells
tells
you
what
kind
of
is
most
important
in
control.
So,
for
example,
if
you're
looking
at
organs
like
the
eye
or
the
liver
or
striated
muscle
or
the
brain
growth
control
is
the
most
important
thing
here.
So
this
is
the
mode
of
control
that's
most
relevant
to
that
scale
or
that
that
type
of
development.
B
Conversely,
if
you're
looking
at
embryos-
and
you
have
like
different
types
of
embryos,
this
is
a
drosophila
embryo,
for
example.
This
is
pattern
formation.
So
actually
it
goes
from
the
embryo
different
types
of
embryos.
Actually,
then,
you
start
to
get
into
the
drosophila
imaginal
disc,
which
becomes
a
wing
in
this
case
and
then
also
in
fishes
like
in
the
phenotype,
when
you
get
like
a
juvenile
phenotype
from
this
embryo
pattern,
formation
is
important,
so
in
the
case
of
this
fish,
striping
is
controlled
by
pattern.
B
Formation,
there's
pattern
formation
in
the
embryo
and
then
there's
pattern
formation
in
one
of
these
organs.
It's
not
growth
control
at
this
point.
It's
just
these
vein
patterns
that
are
being
controlled,
then
there's
a
self-organization
which
is
actually
almost
it's
much
pretty
much
concurrent
with
pattern
formation,
and
this
is
also
where
you
have
a
bunch
of
parts
that
need
to
come
together
and
behave
collectively
or
be
organized
into
a
pattern.
So
that
makes
sense
that
those
two
things
are
overlapping
again.
Self-Organization
is
very
general
term,
but
generally
we
can
use
it
in
this.
B
In
this
area,
then
boundary
organized
is
just
limited
to
the
embryos
here.
Instead
of
these
other
things,
because
boundary
organization
is
basically
defines
the
boundaries
of
an
embryo.
So
if
you
wanted
to
find
the
head
from
the
midsection
to
the
from
the
tail,
you
get
these
boundaries
that
are
set
up
within
the
embryo
and
they
define
tissues.
They
define
different
parts
of
the
what
will
become
different
parts
of
the
organism,
and
so
this
is
actually
overlapping,
with
both
pattern.
B
Formation
and
self-organization,
because
part
of
self-organization
is
where
these
boundaries
are
defined
and
then,
within
those
boundaries
you
get
self-organization
of
different
structures.
Then
there's
a
scaling
to
size,
which
is
it's
smaller,
even
a
smaller
domain.
Yet-
and
this
is
where
things
scale
to
a
certain
size
like
the
head
will
scale
to
the
body
size
and
that
won't
be
maintained
across
development.
B
So
this
feature,
this
is
featured
largely
in
different
types
of
embryos
here
and
then
scaling
to
pattern,
which
is
where
you
get
pattern
formation,
and
then
that
scales
up
with
the
organism
and
then,
of
course,
regeneration
which
is
involved
in
like
say,
striping
or
some
other
set
of
organs.
Not
all
organs,
though-
and
this
is
just
as
regenerative
capacity-
can
you
regenerate
some
the
structure
of
the
organ
or
you
know
something
like
the
pattern
if
it's
damaged
somehow,
and
so
this
is
where
this
comes
into
play.
B
These
are
the
two
modes
of
organization
in
the
control
pattern.
This
is
the
self-organized
case
where
you
get
short
and
long-range
control,
so
this
is
well
known
in
developmental
biology
where
you
get
activation
and
inhibition
depletion
of
different
factors,
so
you
get
different
signaling
molecules,
different
concentrations
of
of
different
morphogens
and
other
types
of
you
know,
gene
products
that
are
defined
in
different
embryos.
B
So
you
get
this
long
range
control
which
is,
over
the
whole
embryo
or
short
range
control,
which
is
locally,
and
you
know
there
are
these
different
circuits
that
people
show
in
papers.
So
this
is
an
example
here
in
this
figure,
where
you
have
this
pulse
of
one
factor
over
the
sort
of
the
anterior.
I
guess
that's
the
anterior
posterior
axis
of
the
organism
and
you
get
these
pulses,
which
might
represent
like
a
a
striping
pattern
of
some
type
and
then
this
red
function,
which
represents
sort
of
what's
co-expressed
with
it.
B
So
you
get
this
pattern
here,
and
so
they
they
combine
in
different
ways
to
define
boundaries,
to
build
these
patterns
or
whatever.
Then
you
get
this
this
pattern
here
at
the
bottom,
which
is
this
dot?
This
dotted
pattern,
which
is
like
a
maybe
an
animal
coat
or
the
surface
of
a
seashell,
and
it
results
from
this
type
of
expression
pattern
and
ultimately,
this
type
of
control
boundary
organized
control
is
where
you
have
all
these
generic
morphogens,
which
are
these
molecules,
but
it's
really
a
theoretical
construct
describing
these
molecules.
B
You
get
transport
of
these
morphogens
and
then
you
get
the
uptake
and
degradation
of
them,
and
so
you
get
these
patterns
where
you
get
these
gradients,
where
you
get
a
concentration
of
morphogen
of
a
certain
type
and
then
it
decays
over
space.
And
then
you
get
another
pattern.
You
get
another
concentration
here
with
a
decay
in
the
other
direction
and
then
there's
a
boundary,
that's
formed
here,
but
you
can
see
that
you
know
those
concentrations
as
they
as
they
migrate
across
or
they
they
diffuse
across
the
cell.
B
They
form
these
stripes
and
it's
just
a
matter
of
constant
overlapping
concentrations
and
the
decay
of
that
concentration.
And
then
that
sets
up
a
boundary
at
some
point
and
you
get
the
striping.
So
these
are
ways
that
you
can
get
self-organization
from
this
kind
of
control.
One
scale
which
is
the
molecular
scale
getting
these
patterns
at
the
morphological
scale.
B
And
then
you
know
there
are
other
types
of
things
that
they
talk
about
with
integral
feedback,
and
this
is
a
very
valuable
paper
and
just
in
terms
of
laying
out
these
different
issues
and
kind
of
going
through
some
of
the
things
in
those
first
two
figures
in
a
little
bit
more
detail.
They
also
talk
about
the
management
of
noise,
which
is
an
interesting
thing
if
you're
interested
in
how
noise
works,
because
noise
is
actually
very
important
in
a
lot
of
biological
systems.
B
B
F
B
I
didn't
see
it
too
explicitly,
but
I
mean
this
is
just
like
examples
of
this,
so
they're
just
kind
of
hitting
examples
from
the
literature,
but
it
would
definitely
play
a
role
if
you
think
about
like
the
different
mechanisms
and
they're,
probably
mechanisms
we
don't
even
know
exist
to
control
some
of
these
things,
but
you'll
draw
out
those
circuits
and
figure
out
like
what
are
all
the
factors
involved
in
this.
You
know
and
again
it's
different
for
different
systems,
so
yeah.
F
B
C
C
B
Like
focused
on,
like
the
molecular
scale
in
systems
biology
like
they're,
interested
in
like
metabolic
networks
or
gene
expression
networks,
and
you
don't
see
a
lot
of
papers
on
like
physical
regulation
in
in
systems
biology
in
in
developmental
biology,
you
probably
will
see
that
more.
But
that's
where
they're
borrowing
that
language
from
it's
kind
of
coming
from
control
theory
and
then
systems
biology
where
it's
been
modified.
A
bit
for
biological
systems,
but
not
every
type
of
biological
system
and.
B
Yeah
well,
it
was
kind
of
like
a
survey
paper,
but
it
was
like
taking
those
concepts
and
rooting
them
in
developmental
biology.
So
it's
like
you
know.
He
goes
through
a
bunch
of
different
concepts
from
what
are
essentially
things
that
systems
biology
would
see,
and
people
read
about
like
all
the
time
but
then
bring
them
to
a
developmental
biology
context
where
people
maybe
aren't
thinking
about
them
all
the
time.
And
then
you
know
it's
just
kind
of
that
transition,
but
yeah,
it's
a
nice.