►
Description
DevoWorm meeting: January 20, 2020. Attendees: Susan Crawford-Young, Richard Gordon, Vinay Varma, Devansh Batra, Yash Agarwal, Bradly Alicea, and Jesse Parent
A
B
A
B
A
A
B
B
B
B
A
B
A
Working
on
a
couple
of
different
papers
and
did
the
vassal
area
paper-
and
we
did
a
couple
of
other
finishing
up
some
papers
on
Brayton,
Berg
vehicles
and
things
like
that.
So
it's
been
a
pretty
diverse
collection
of
things
do
differ,
you
know,
keep
up
on
different
things
and
they
do
a
lot
of
shifting
back
and
forth
between
like
different
topics,
so
hi
dick.
C
A
All
right
so
welcome
to
the
meeting.
We
don't
have
anything
we
have
a
couple
of
things
planned,
but
we
don't
have
any
talks
today.
Like
long
talks,
we
have
talked
a
little
bit
about
gee
sock,
the
equals
Summer
of
Code,
and
then
Susan
is
going
to
talk
about
her
the
paper
that
she,
you
know
we
profiled
last
late
last
year
on
embryos
and
she'll
talk
about
what
in
more
detail
about
that
paper.
A
A
Yeah,
it's
fine!
Why
don't
I
start
with
the
G
sock
stuff?
They
present.
So
a
little
bit
of
news
on
this.
The
G.
Some
proposals
are
up
on
high
n
CFS
website.
Let
me
show
you
this
so
I
n
CF
is
now
running
narrow
stars
and
something
called
narrow
stars.
It's
based
on
on
discourse.
For
those
of
you
know
with
discourses
it's
it's
kind
of
like
slack,
but
it's
the
public
forum
that
has
it.
They
have
a
special
software
that
people
can
set
up
their
own
discourse.
A
So
knurl
stars
is
one
of
those
discourse
platforms
and
I
ncf
is
using
it
for
a
lot
of
interactive
communication
for
a
lot
of
the
people
and
I
n
CF.
So
there
are
a
lot
of
resources
here.
First
of
all,
they're
educational
materials
for
neuroscience,
there's
a
community
encyclopedia
and
their
standards
and
best
practices
and
they're
all
and
hosted
on
neuro
stars
and
people
can
comment
on
them
and
things
like
that.
One
of
the
other
things
that
neuro
stars
is
hosting
are
the
ji-suk
project.
So,
where
I
see
you.
A
This
is
through
open
worms.
So,
if
you
go,
you
know
this
is
sort
of
will
have
the
community
period
through
open
worm,
which
is
you
know,
to
interact
with
all
the
different
groups
in
open
worms.
So
you
know
for
this
project
you,
you
know
there
would
be
like
three
weeks
where
you
present
to
the
main,
open
worm
group
and
then
you
know,
you'd
have
people
from
all
the
different
sub
projects.
A
You
know
getting
exposure
or
what
you're
going
to
be
doing
and
then
you
might
make
some
connections
there.
You
know.
Usually
we
do
the
community
facilitation
period
over
slack
and
then
you
get
into
the
coding
after
a
few
work
weeks,
and
so
I
have
a
couple
of
enquiries
here,
here's
a
one
inquiry
already
they're
just
kind
of
introducing
themselves
and
then
you
know
kind
of
getting
them
in
touch
with
resources.
So
the
first
stage
of
this
project
is
to
have
people
submit
proposals
for
the
project.
A
So
you
know
here's
the
description
and
some
people
are
asking
some
more
specific
questions
about
how
to
go
about
applying
and
then
once
we
get
proposals
in
and
you
know
they
go
through.
Maybe
a
couple
of
rounds
of
editing
on
those
then
we'll
get
to
evaluate
those
proposals.
So
that's
for
pre-trained
models
and
again
it's
just
you
know
matter
of
them.
Reading
through
this
proposal.
A
That
idea
15
is
the
open,
Divo,
so
integration
and
of
course
this
is
the
one
that
we've
discussed
in
the
meeting
that
follows
up
on
Vinay's
work,
and
this
is
the
description
for
that
again.
It's
they
go
through
the
description
and
they
can
ask
questions
so
I
have
an
important
your
ebony
answered.
Just
you
know
they
just
people
inquire
about.
You
know
whether
they'd
be
a
good
fit
and
usually
they
are.
But
you
know,
then
you
have
to
kind
of
steer
them
in
the
right
direction
in
terms
of
putting
together
a
proposal.
A
Comments
here
then,
a
I
went
through
the
ji-suk
ideas
and
they're
interesting
I,
guess,
project
14
and
15
are
related
to
open
work.
Yes,
I,
don't
know
I,
don't
think
open
lung
has
given.
Doesn't
you
know
there
are
no
other
projects
associated
with
open
room
this
year,
I
made
a
couple
calls
and
no
one
seemed
interested.
So
I
don't
know
if
anyone's
going
to
submit
I
think
you
have
until
the
end
of
the
month
is
submit.
A
project
is
40
montaging
of
axolotl
women
yeah.
A
A
Is
the
formation
of
well
I
mean
we
can
talk
about
it?
That's
probably
too
late
for
this
year
to
get
something
together,
because
I
like
to
develop
it
a
little
bit,
but
probably
next
year,
we
can
do
like
something.
Definitely
I
think
we
can
do
it.
Oh
just
you
have
to
figure
out
what
the
tasks
are
and
everything
so.
A
Okay,
so
open
a
said,
he
gave
me
some
leads
and
asked
me
to
contact
them.
Yeah
I,
think
the
person
in
this
person
here
I
think
was
from
today's
University,
so
he
did
get
there.
Are
people
were
kind
of
feeding
into
that?
Okay,
yeah
Jesse
says
I
want
anything
to
present
today,
but
I
wanted
to
stop
in
well.
Thank
you
for
showing
up
Jesse,
so
yeah.
A
B
A
A
So
Jessie
I
see
we're
working
on
the
paper
this
weekend.
So
thanks
for
that
and
that's
been,
you
know
trying
to
find
the
time
for
that.
So
yeah,
that's
that's
fine!
Then
you
had
some
other
ideas
about
what
you
want
to
do
present
to
the
group.
So
yeah,
like
the
one
you
sent
me,
you
sent
me
a
couple
ideas
where
they're
all
fine
to
do.
We
can
do
a
short
presentation.
B
B
A
B
So
there's
the
reference
and
what
you're
supposed
to
know.
This
is
all
these
first
bubbles
here
I've
had
a
car
on
what
stage
you
need
stage,
32
more
just
before
our
boss.
This
thing
starts
to
form
and
what
they're
saying
is
that,
because
of
the
high
pressure,
this
is
like
fracking
when,
in
between
the
cells?
B
B
B
Maybe
I
had
a
couple
weeks
ago,
but
I
I've
been
through
some
exams
and
stuff,
so
I
just
sort
of
again
it's
a
copy-paste
yeah,
so
it
says
multi
scale,
feedback
interactions
between
low
pressure
and
tissue,
mechanics,
which
I
will
blast.
This
is
size
so
they're
saying
that
that
this
country
is
this
the
size
of
this
blast
assisted.
B
In
other
words,
this
thing,
this
anvil
pasta,
has
the
hatched
we're
hosting
hatching.
Contractions
and
I
wrote
a
paper
about
this
quite
a
long
time
ago,
and
they
said
that
a
young
mothers
they
had
too
many
contractions
and
the
old
mothers
just
like
any,
and
it
would
crack
the
zona
pellucida
and
it
would
hatch
hatch
out
and
here's
a
picture
of
that
Mouse.
Oh,
this
is
quite
a
wide
crack,
but
usually
it's
trying
to
slip
out
of
a
very
tiny
area.
I
guess
I
should
have
gotten
one
for
contrast,
but
anyway,
this
is
a.
B
B
Anyway,
so
so
part
of
what's
going
on
is,
if
you
have
string,
no
catalysts
work
better
and
chemical
processes
work
better
on
a
curve.
That's
true
of
the
mitochondrial
ease
and
it's
the
truth
embryo.
So
if
you,
if
you
have
this
type
of
thing,
going
on,
you've,
you've
got
more
chemical
reactions
happening.
B
Yeah
the
strain
and
the
pressure,
what
a
name
enable
different
types
of
chemistry,
ISM
and
then
this
kind
of
looks
like
it's
soft
state,
a
splitter,
and
so
you
get
different
salad
Heejun
when
and
so
where
this
is
going
on
again.
I
haven't
gotten
back
to
to
us.
So
I
haven't
completed
my
thought
about
this,
but
there's
definitely
active
myosin
and
microtubules
and
play
with
this,
and
this
is
another
slide
showing
that
so
actin
myosin
Network
for
force
generation.
B
B
And
this
is
another
diagram
of
how
a
song
becomes
a
moving.
So
little,
then
this
will
I
tear
this
one
in
here,
because
it
looked
interesting,
but
I
haven't
got
to
the
bottom
of
it
yet.
So
if
this
is
going
to
be
very
brief
and
then
this
is
actually
some
work,
this
thing
with
doctor,
Sharif
and
I
took
this
image:
harsh
a
religion
era
and
aid
offices
stage
and
then
again
in
his
his
grad
student,
took
this
image
and
I
believe
it's
supposed
to
be
more
of
an.
B
B
I'm
a
blast
assume
for
the
same,
and
you
can
see
there's
a
should
be
extreme
stress
here
when
you
believe
this
is
under
a
lot
of
pressure
as
well
yeah,
so
this
is
sort
of
backwards
to
or
mammalian
and
or
this
is
an
amphibian
embryo
and
late
into
the
process
feel
inside
and
the
animal
development
on
the
surface
or
on
the
surface
here
instead
of
being
internal,
but
again
pressure
isn't
out.
So
it
looks
like
it
is.
So
a
part
of
the
task
is
to
measure
the
pressure
inside
inside
of
the
pasta.
B
A
B
B
B
B
E
B
B
A
B
B
Yeah,
that's
I
have
to
learn
about
it
here,
but
definitely
he's
got
a
set
up
so
that
you
can
see
what's
going
on,
it's
not
as
refined
as
maybe
I
would
like,
but
they're
trying
different
things
in
the
lab.
This
is
the
one
right
here
is
a
quadrature
Lucy
key
I
believe
and
it's
a
swept
source
source
quadrature.
E
B
Yeah,
so
it's
perturb
it
maybe
with
ultrasound
I,
don't
know
so
it
it
shakes
the
bends
or
you
compress
it
a
little
bit
and
then
it
it
changes
change
of
shape
the
air-
and
you
can
measure
the
before
and
after
pictures,
to
see
how
much
advance
to
the
II
am
a
starter.
Fluid
viscosity
are
the
last
order
feet
together,
according
to
some
of
the
papers
that
I've
been
reading,
yeah
yeah.
B
A
B
B
A
G
E
A
E
E
B
E
B
A
B
A
For
presenting
it
was
good
if
I
guess,
if
people
want
to
see
the
videos,
you
can
go
online
to
the
paper,
the
usual.
We
have
the
supplemental
materials
there
for
people
to
view,
and
so
then
I
guess
the
next
question
I
have
for
people
is
there
are
people
interested
in
presenting
things
in
coming
weeks?
So
we
have
our
schedule
I
know
next
week
and
about
you're
going
to
be
presenting
on
I
can't
remember
what
the
actual
topic
is,
but
they're
gonna
be
presenting
something
machine-learning
oriented
and
we'll
see.
A
You
know
how
that
fits
into
all
this
I
mean
you
guys
can
present
and
then
we'll
ask
questions
and
that'll
be
interesting.
I,
don't
know
where
two
bunches,
who
maybe
left
early
but
and
so
that's
on
the
schedule
for
next
week
and
then
the
week
after
that,
I'm
not
sure
I,
know
Jesse
wants
to
present
sometime
soon
on
some
things.
A
You
know
we
can
schedule
that
we
can
in
schedule
it
offline
even
like
at
another
time
of
the
day
and
then
I
we
can
replay
okay.
Lunch
is
still
here
and,
like
I
said
for
Jesse
we
can
present,
you
can
present
at
another
time
if
it's
more
convenient
and
then
we
can
play
the
video
of
that
in
this
meeting.
Just
give
you
know,
so
you
know
make
it
more
convenient
for
you,
okay,
yeah,
the
replay
thing
that
would
work,
and
you
gave
me
a
couple
of
ideas
on
different.
C
A
And
things
that
you
have
venture
also
I
suggested,
maybe
I,
don't
know
I
mean
you
really
want
to
attended
the
the
Gary,
Marcus
and
Ben
geo
to
be
I.
Don't
know
if
you
feel
like
you
want
to
give
a
short
recap
of
that
at
some
point.
Yeah,
because
I
mean
that
was
interesting,
sort
of
a
you
know,
I
think
there's
a
lot
of
interpretation
needs
to
be
done
on
that.
So
that's
something
to
think
about,
and
then
the
name
might
be
interested
in
presenting
something
there.
Anything
you've
seen
that's
interesting,
Vinay,
okay,.
A
For
the
second
blog
post
and
pre-trained
models
so
yeah
we
actually
go
easily
talked
about
that
before
Christmas.
That
was
the
post.
That
I
don't
remember
what
the
actual
topic
was,
but
yeah
there's
probably
some
follow-up
to
do
on
that
and
again
like
if
you
want
to
propose
a
blog
post,
okay,
an
extension
of
the
first
blog
post
I.
Think
that's
good!
Well!
A
So
if
you
want
to
do
that,
then
you
can
send
me
yeah,
maybe
a
draft!
Maybe
an
outline
is
more
appropriate
and
then
we
can
talk
about
it
meeting
you
know
like
in
Google
Docs
or
something
yeah
this
weekend.
Just
just
outline.
You
know
at
some
points
that
you
want
to
hit.
You
know
you
know.
What's
what's
what?
How
do
you
want
to
extend
this
first
post?
A
Think,
especially
given
the
summer
code
present,
eight
or
the
Summer
of
Code
proposals
that
are
gonna,
be
written,
I
think
having
that
second
post
would
be
good
sort
of
guide
people
on
them
and
then,
of
course,
we
can
also
do
things
that
are
more
developmental.
So
we've
mentioned
that
we're
going
to
be
getting
the
axolotl
data.
So
there's
the
status
of
that
Richard
and
Susan
have
been
developing.
A
A
You
present
it
last
week
or
last
year,
where
they
flipped
the
microscope,
the
invert,
basically
they
invert
the
embryo
when
it's
under
magnification
and
you're
able
to
see
you
know
the
embryo
in
different
at
different
orientations,
and
so
Susan's
going
to
be
sending
me
those
data
and
then
we'll
be
trying
to
figure
out
how
to
best
represent
the
data.
You
know
there
I
mean
a
lot
of
different
ways.
Computationally
to
deal
with
it
does
axolotl,
it
says
X
a
lot.
It
was
a
lot
like
Caribbean
ancestors,
there's
also
early
Caribbean
ancestors.
A
I
hope
that
they're
a
lot
more
freeform
than
last
semester
or
last
year,
where
we
did
the
presentations.
You
know
we
kind
of
had
a
presentation
every
week,
but
I
hope
that
we
can
also
get
some
good
ideas
going,
so
maybe
some
new
papers
going
then
there's
also
a
follow-up
paper
to
the
to
the
festival
area
paper.
So
the
best
wary
of
paper,
the
first
one
was
the
machine
learning
analysis
of
the
or
follow
G.
A
We
were
thinking
of
doing
a
second
bachelor
area
paper,
I'm,
a
soul
area
behavior
that
one
still
kind
of
in
the
thinking
stages
right
now.
So
maybe
we
can
come
up
with
an
outline
of
that
for
maybe
next
week
or
the
coming
week
after
that,
like
I
talked
a
little
bit,
he
mailed
a
little
bit
with
Thomas
Harvick
about
that
he's.
Actually
cultured
Bassel
area
and
he's
captured
a
lot
of
microscopy
data
on
it,
but
we're
still
kind
of
thinking
about
that.
A
So
in
a
dick
was
had
proposed
on
the
aegis
paper
and
I
thought
it
was
a
good
idea.
So
you
know
again.
This
is
something
that
we'll
be
developing
over
the
next
coming
week,
so
if
anyone
else
has
ideas
about
that,
if
if
you're
interested
in
behavior
in
an
organism
doesn't
necessarily
have
any
kind
of
brain,
we
can
talk
more
about
that.
Oh
we
have
some
things
in
the
chat
here.
A
Jesse
says
I'm,
not
sure
if
embodied
cognition
would
be
better
for
this
group
of
the
other
one.
But
I
may
do
something
on
that
general
definition,
perhaps
like
I'm
working
on
the
manuscript
I.
Think
probably
the
other
group
would
be
better
for
that
at
least
initially.
So
we
could
do
that
in
the
other
group
and
then,
if
it's
it's
really
interesting
to
this
group,
we
might
talk
about
it
here
too,
and
then
Yash
says
wanted
to
clarify
the
we'll
be
talking
about
some
DL
architectures.
That's
deep
learning!
A
Architectures
next
week,
capturing
spatiotemporal
features
from
videos
and
sets
of
images,
so
it
may
be
useful
for
the
pre-training
models
yeah.
That
would
be
very
good
if
you
could
tie
it
into
the
pre
train
models
or
if
you
could
remind
us
that
it
does
tie
in
to
that
at
least
that
would
be
a
good
discussion.
Point
I
think
you
know
I,
don't
know
you
know,
I,
don't
know
much
about
the
architectures
like
that.
I
can
like
converse
about
right
now,
but
yeah.
A
So
you
know
when
you
present
make
sure
you
know
make
sure
that
we're
you
know
as
a
reminder
room
mixed
group.
We
have
people
who
are
computer
science,
people
and
people
who
have
different
backgrounds,
learn
as
computer
science.
We
rented.
So
you
know
try
to
if
there's
jerk
and
try
to
make
sure
that
you
sort
of
you
know
explain
it
make
sure
you
leave
time
to
explain
it
to
people
or
you
know,
make
sure
that
it
isn't
very.
Just
like
you
know,
cutting
and
pasting
from
a
paper.
A
A
So
dick
said
here
in
this
comment:
Basso
Larry
needs
simulations
of
alternative
mechanisms,
so
this
is
for
a
behavior,
so
yeah
for
behavior
that
yeah.
There
are
a
lot
of
different
things,
different
ways.
We
can
look
at
the
behavior
and
sort
of
what's
going
on
our
very
generating
that
behavior
and
then,
where
we
would
match
it
to
movies
or
data
of
the
vessel
very
actually
moving
and
so
I
don't
know
people
are
familiar,
but
in
the
paper
Thomas
does
an
analysis
where
he
shows
movement
of
the
basilar
area
colony
there.
A
These
rod
like
cells
that
are
arrayed
in
a
stack
like
this
and
they
move
in
a
way.
That's
you
know
they
move
laterally
to
one
another,
so
they
generate
these
movement
patterns
that
are
quite
exquisite
for
something
that
seems
like
a
pretty
simple
organism.
You
know
they
fall
kind
of
like
an
accordion.
They
move
laterally.
A
Their
edges,
and
so
they
generate
these
sinusoidal
waves,
but
it's
not
just
as
simple
as
a
sinusoidal
wave.
There
is
a
lot
of
variation
in
there,
and
so
we
already
know
what
the
mechanism
is.
That
coordinates
the
movement,
but
that's
kind
of
the
idea,
this
papers
that
we
would
be
exploring
that
aspect
of
it.
But
we
have
to
you
know,
keep
in
mind
that
you
know
we
have
to
validate
the
data.
Somehow
so
I
mean
they
have
to
basically
be
alternative
mechanisms,
but
they
can't
be
that
outlandish.
A
That's
good
I
think
we
have
a
couple
of
things
that
we'll
be
working
towards,
and
hopefully
everyone
is.
If
anyone
has
any
questions
you
can
contact
me
in
slack.
You
can
email
me
or
you
know
just
wait
until
next
week.
I
guess,
if
you
want
to
talk
about
something
during
the
meeting.
Just
let
me
know
a
little
bit
in
advance,
but
otherwise
you
know
I
think
next
week.
So
we
have
the
D
architectures
presentation,
we'll
have
a
lot
of
time
for
anything
else.
D
A
How
long
do
you
think
the
presentation
will
be?
Maybe
we
should
probably
aim
for
about
4045
minutes.
This
is
usually
a
good
time
for
presentations.
You
know
we
can
make
them
shorter,
but
I
just
want
to
yeah.
40
minutes
would
be
good,
so
40
minutes
will
be
good.
Then
we
can
discuss
for
about
20
minutes.
A
Like
Susan's
presentation
was
pretty
good.
She
had
you
know
she
gave
some
images
and
you
know
a
little
bit
of
background
in
it.
You
know
so
now
we
have
a
good
taste
of
what
that
research
looks
like
and
then
you
know
you
can
also
go
into
more
detail
presentations,
but
you
know
it's
like
it's
up
to
you
so
well,
anyways
a
thank
you
for
meeting
this
week.
It's
been
fun
and
next
week
we
will
talk
about
deal
architectures
and
again.
If
you
have
any
thing
you
want
to
propose
for
some
future
week.