►
From YouTube: Open Source Antibiotics Science Update July 23rd 2021
Description
Weekly open project meeting for Open Source Antibiotics Series 2.
Full Project: https://github.com/opensourceantibiotics/Series-2-Diarylimidazoles
Relevant GitHub Issue: https://github.com/opensourceantibiotics/Series-2-Diarylimidazoles/issues/83
On the call: Professor Matthew Todd, Dr Dana Klug, Dr Edwin Tse, Giada Sabatino, Alex Vaideanu (UCL), Prof Lee Graves (UNC Chapel Hill), Dr Flavio Emery (University of Sao Paulo), Dr Quillon Simpson (NEU), Anthony Sama (Citizen Scientist), Dr Chris Swain (Cambridge Medchem Consulting)
A
That's
all
right
all
right,
hi,
everyone,
friday,
23rd
of
july,
open
source,
antibiotics,
series,
two
and
and
I'm
gonna
share
the.
A
Relevant
thing
now,
the
first
thing
I
was
just
going
to
mention
before
I
forget
is
that
yeah
holidays
are
coming
up
right
and
our
resident
potency,
tester,
paul
stapleton,
is,
is
going
to
be
doing
some
vacation
and
essentially
is
able
to
do
another
screen
for
potency
of
compounds.
A
If
we
get
them
to
him
asap,
and
by
that
I
mean
like
monday
tuesday,
so
we've
got
compounds
that
came
in
from
northeastern,
which
is
fantastic.
Thank
you
so
much
for
sending
those
in
and
of
course,
we've
got
some
local
compounds
as
well
from
dana
and
from
and
giada,
and
we
may
potentially
have
combats
from
flavor,
but
it
depends
how
it's
going
but
is.
Is
it?
Is
it
kind
of
worth
it
and
reasonable
to
stick
those
in
for
an
assay
essentially
next
week,.
C
Don't
you
have
the
compound
that
has
the
totally
changed
from
the
peridial
to
something
like
thiazole
right,
yeah.
B
I've
got
the
thiazole
and
the
pyrrole.
D
A
Of
those
would
be
good
to
go
on
like
monday,
tuesday
kind
of
thing.
A
All
right,
let's
do
that
great
I'll
I'll,
send
the
mail.
I
just
spoke
with
him
yesterday
and
I'll
send
the
mail
to
him
after
this
meeting
to
say
that
we
think
there'll
be
some
compounds
coming
in
that's
great.
He
yeah
he's
he's
got
a
bunch
of
stuff
before
he
needs
to
do
it
before
he's
away
on
vacation,
but
he's
in
the
school
next
week
and
can
do
some
testing.
A
Oh
yeah,
that's
good,
but
then
we'll
have
like
a
three
week.
Hiatus,
I
think,
because
august
is
is
august.
A
Okay,
that
would
be
really
good.
So
that's
the
first
thing.
Yeah
second
thing
was
so
this
this
archive
thing
about
benjamin
is
going
to
dndi.
We
we
keep
missing
that
out.
We're
not
doing
that
anymore.
Are
we
doing
that.
A
B
Yes,
I
emailed
lori
about
this
actually
and
you're
happy
to
have
them
right.
Laurie
yeah
and
I
I
did.
E
E
It
yeah
we've
emailed
about
this.
It's
happening.
We
sent
files
to
you
in
our
last
shipment,
so
yeah
we
should
be
able.
We
should
be
good
to
go
now.
B
Yeah
we
haven't
put
that
together
yet
because
we've
been
trying
to
get
jada
finished
up
mostly,
I
think,
but
hopefully
we
try
to
shoot
for
next
week,
maybe
again
yeah.
E
Awesome
how
many
compounds
are
coming
in
for
dndi?
Do
you
know
roughly.
B
A
Thanks
and
then
speaking
of
the
parallel
dndi
assay,
so
then
you
very
helpfully
posted
the
updated
color-coded
set
of
data
and
calculated
some
cell
activities,
all
of
which
look
red
pretty
much
indicating
not
great
selectivity,
but
there
are
some
oranges
where
there
is.
There
is
a
bit
of
a
window
depending
on
which
data
you
get.
I
guess
you
prioritize
and
there's
been
some
discussion
here
about
the
value
of
some
of
these.
Some
of
these
compounds
laurie.
A
E
Yeah,
so
I
guess
a
couple
of
things:
if
you
could
scroll
back
up
just
for
a
second
to
the
sis,
so
for
the
assays
themselves,
t
cruisy
uses
mrc5
and
ln
phantom
uses
the
pmm.
E
So
those
are
like
those
are
the
two
relevant
kind
of
columns
for
each.
We
typically
would
be
aiming
for
a
selectivity
window
of
about
10.,
so
I'd,
say,
dean
is
probably
a
little
on
the
pessimistic
side
with
some
of
the
color
coding.
You
know.
Obviously,
the
plus.
E
I
guess
my
my
thinking
on
this
is
in
terms
of
the
assays
themselves.
I
would
probably
be
focusing
for
our
purposes
like
right
now
on
the
actual
mrc,
5
and
pmm
values
in
the
in
the
comment
above
this,
because
that's
really,
you
know
we're
not
so
fussed
about
the
selectivity
index.
It
was
more
about.
You
know
what
is
the
the
potency
versus
in
mrc5
and
in
pmm
for
some
of
these
compounds,
so
that
was
kind
of
where
I
was
going
from
that.
I
also
think
it's
it's
kind
of
interesting.
E
So
if
you
compare
osa
865,
which
is
the
third
one
in
on
the
top
row
and
osa975,
which
is
the
second
one
from
the
right,
like
all
you've
changed,
there
is
a
methyl
to
a
nitrile
and
significantly
changed
and
modulated
the
the
tox
against
pmm
and
mrc5,
and
so
I
wonder
actually,
if
what
we
should
do
is
as
kind
of
more
of
an
in-depth
analysis
like
that
thinking
about
okay.
E
What
do
we
know
about
the
the
sar
and
then
what
do
we
know
about
the
top
side
of
things
and
and
whether
we
can
actually
glen
in
this
way
now
that
we've
kind
of
got
a
bit
of
a
set
that
was
kind
of
where
I
was
coming
at
it
from
and
the
other
thing
is
we
cautioned
a
couple
of
months
ago
to
really
just
be
focused
on
in
vivo
toxin?
If
in
vivotax
is
an
issue
then
go
forward
acknowledging
that
these
compounds
are
slightly
different?
E
We
progress
the
n
alkyl
derivatives
into
pk
and
there
was
no
tox
there.
You
know,
maybe
it's
not
so
terrible.
E
A
A
I
can't
it
was
against
last
I
checked,
it
was
against.
You
know:
60
different,
supported
cell
lines
right
unless
someone's
looked
at
it
more
recently
than
I
have,
which
is
supposed
to
give
you
information
on
types
of
talks.
Isn't
it
as
far
as
I'm
aware
or
potential
mechanisms
of
action
and
things
like
that
or
mechanisms
of
talks?
That's
that's!
As
far
as
I
am
aware,
so
you
know
a
super
version
of
this
against
different
cell
lines
that
we
may
not
have
access
to.
A
I
guess
I
haven't
actually
checked
if
these
are
part
of
that,
but
that,
but
that
aside
so
lori
are
you
I
mean
from
the
data
in
front
of
us
here.
Are
you
more
interested
here
in
in
in
sort
of
pursuing
compounds
with
decent
potency
against
the
the
pathogen,
or
are
you
interested
in
trying
to
just
find
a
pattern
that
allows
us
to
modulate
the
talks.
E
I
think,
both
from
our
perspective
so
to
to
so,
as
dana
mentioned
in
her
comment
for
tea,
cruisi
and
leash,
it's
incredibly
challenging
to
actually
find
a
heat
compound
and
for
us,
a
hit
compound
is
anything
under
10
micromolar
right.
So
a
number
of
these
are
actually
already
submicromolar
and
at
that
point
you're
actually
kind
of
getting
into
serious
sort
of
levels
of
potency.
E
Yes,
tox
the
selectivity
index
then
does
become
relevant
and
some
of
the
some
of
these
it's
it's
not
amazing,
but
we
actually
think
that
there
could
be
a
path
forward
in
terms
of
maybe
bringing
in
some,
maybe
modulating
talks
and
potency,
and
so
quillin
has
actually
pulled
together
about
compounds
that
maybe
we
could
make
with
a
view
towards
sort
of
our
focus
which
has
been
on
the
parasite
side
of
things
and
so
he's
actually
shared
that
list
with
dean.
I'm
not
sure
if
you've
put
it
on
the
github
yet
colin.
G
Yeah,
so
I
I
briefly
put
that
up
last
week
in
the
meeting
it's
in
the
slides
that
I
attached.
E
So
for
us
there's
a
couple
of
strategies
that
we
could
use.
One
would
be
to
actually
sort
of
pursue
this
series
a
little
better,
a
little
more
for
the
the
crazy
leash
aspect
and
then
the
other.
It
would
be
actually
to
have
a
look
at
the
data
and
see
if
we
can
understand
the
structure,
toxicity
relationships.
A
G
Yeah
bear
with
me:
okay,
no
problem.
So,
to
be
honest,
it
may
be
quicker
if
you
just
pull
up
those
slides
okay.
So
these
slides
here,
yeah,
okay,
all
right
I'll,
do
that.
D
H
G
So
so
that's
the
the
series
of
compounds
that
I'm
I'm
currently
going
through
and
making.
The
idea
was
sort
of
looking
very
briefly
at
the
sar
for
the
compounds
that
we
got
through.
G
We
saw
a
preference
for
substitution
in
three
position,
so
so
I
was
just
sort
of
looking
in
through
the
lens
of
trying
to
further
explore
some
of
the
best
compounds
that
were
substitute
in
the
fourth
position,
like
the
trifluoromethoxy,
for
example,
and
then
putting
them
into
three
positions
to
see
if
we
can
get
improved
tox,
which
is
what
occurred
when
you
had
the
methyl
scan
going
around
between
the
two
three
and
four
position,
the
other
component
for
this
series
that
laurie
mentioned.
G
We
we
have
gone
into
pk,
but
have
failed
to
get
suitable
exposure
for
the
compound.
So
that
was
for
the
n
alkyl
compounds
where
we've
gone
through
with
the
pk.
So
a
lot
of
our
optimizations
have
been
really
trying
to
nail
in
compounds
that
are
more
metabolically
stable,
as
well
as
as
soluble
in
the
actual
assay
as
well.
G
So
that's
sort
of
why
why
we
sort
of
looked
at
these
compounds
with
with
the
lens
towards
lipophilicity,
even
though
I
did
say
I
think,
just
above
where
I've
attached
the
slides
here,
that
there
isn't
a
strong
correlation
between
the
felicity
and
solubility.
For
this
series.
G
A
A
So
we
need
to
keep
on
sharing
compounds
right.
I
mean
that's
the
take
home
here,
I
suppose,
and
keeping
an
eye
on
this
because
yeah.
I
agree
that
some
of
these
fairly
straightforward
compounds,
particularly
with
the
meta
on
the
ring,
have
promised
I
mean,
according
to
sub
micro,
molar
beam
of
great
promise.
Then
then
yeah
we've
got
some
nice
compounds
here
for
sure.
C
E
It's
inactive,
oh,
okay,
yeah
look!
I
mean
these
are
completely
different
organisms.
It
doesn't
shock
me
that
we're
seeing
so
I
I'm
I'm
not
super
surprised
by
that.
E
Yeah
yeah,
I
mean
I
I
agree,
but
there's
also
compounds
in
this
set
that
are
even
more
potent
that
I
think
are
a
little
bit
more
interesting
at
this
point.
E
We
only
get
in
vitro
intrinsic
clearance
data.
E
So
so
we
do
have,
we
do
have
access
to
a
cake
or
two
assay.
Can
you
remind
me
of
the
top
of
your
head?
Have
any
of
that
and
now
calls.
G
A
H
Well,
the
other
thing
would
be
if
you
have
a
mdk
cell
line
with
transporters
and
things
like
this,
and
it
might
give
you
more
information
about
what
the
issue
is
with
absorption.
If
there
is
one
rather
than
just
solubility,.
A
A
Okay,
would
you
could
you
just
dump
that
as
a
comment
and
the
issue
for
today's
meeting,
just
to
make
sure
we
don't
miss
it.
A
All
right,
I'm
just
conscious
of
the
time,
because
I
have
a
stop
at
half
past.
Unfortunately,
and
I
just
want
to
make
sure
that
we
deal
with
anything
else
that
we
need
to.
A
So
we
are
so
that
the
talks
is
still
going
on
in
the
background
and
I'm
sure
alex
will
update
it
soon
when
she's
completed
the
table
that
she
talked
about
last
time,
current
synthetic
targets-
I
guess
we
talked
about
those
relatively
recently
and
I
don't
think
that
flavio
is
here
to
tell
us
about
the
homologous
series,
unless
I'm
mistaken
so
and
giada
who's
on
the
call
has
her
last
day
today
in
the
lab,
she's
writing
up
thesis
and
stuff.
A
It's
the
last
thing
in
the
lab
and
it
looks
like
she's
managed
to
make
another
compound,
which
is
very
exciting.
So
I
said
that
I
would
share
a
slide
and
I
don't
have
it
so
you
just
talk
amongst
yourselves
for
like
two
seconds.
A
F
All
right:
well,
there
it
is
yeah.
I
I
don't
know
exactly
what
you
can
see,
but
if
you.
F
Okay,
okay:
this
is
the
compound
and
it's
kind
of
similar
of
the
one
that
I
did
last.
We
did
last
week
during
the
last
weeks.
I
just
added
some
preparation
here
and
I
have
44
milligrams,
so
yeah.
F
F
A
Right
and
your
nmrs
are
beautifully
pure,
like
in
a
textbook.
F
F
F
A
All
right
really
good,
so
perfect.
Anyone
else
wanted
to
give
a
a
nano
chemistry
update.
A
B
Just
straight
onto
a
reverse
face
column:
I
don't
know
how
I
feel
about
that,
because
I
think
that
it
will
stain
the
column,
but
I'm
gonna
try,
I'm
not
seeing.
I
don't
think
the
same
impurities
that
you
were
seeing
quillin,
so
I
might
just
try
crashing
it
out
of
the
dmso
solution
and
see
how
that
goes.
But
I
need
more
starting
material
in
to
make
the
next
analog.
So
I
can
try
that,
but
hopefully
will
work.
G
I
was
just
going
to
say
that
if
you,
if
you
find
the
columns,
are
getting
stained
by
the
isocyanide
byproducts,
I
flush
it
with
about
if
it's
a
30
gram
column
with
about
300
ml
of
isopropanol
and
that
sort
of
freshens
it
back
up
ready
for
use.
E
I
was
actually
going
to
suggest,
I
wonder
if
you
could
even
use
like
the
little
sample
loading
cartridges,
whether
that
would
then
just
trap
the
staining
on
like
the
loading
cartridge,
and
you
could
then
just
turn
it
on
flush.
B
A
Okay,
great
and
then
just
the
last
thing
I
was
going
to
mention
is
that,
on
the
basis
of
last
meeting,
I'm
I'm
just
hassling
lee
about
whether
he's
got
enough
compound
to
do
the
next
mechanism
of
action
experiments.
That
was
so.
That
was
all,
and
I
I
posted
about
whether
or
not
if
we're
interested
in
this
proteasome
idea,
which
is
you
know,
I
mean
it's
a
crapshoot,
isn't
it
sort
of
naming
a
target
and
seeing
what
happens?
But
if
we're
interested
in
that
it
sounds
like
we
use.
A
I
was
interested
in
whether
a
lab
was
around.
That
could
run
an
essay
that
might
suggest
whether
or
not
I'm
compensating
the
pretty
zone,
and
there
is
one
in
karolinska
and
I
was
going
to
reach
out
to
the
guy,
who
runs
that
and
see
what
happens
if
anyone's
getting
other
bright
ideas.
Then
great,
it's
a
you
know
it's
a
chance,
but
you
know
I.
C
Mean
because
we
really
don't
know
because
if
you
run
it
through
the
general
machine
learning
stuff,
it's
so
similar
to
kinase
inhibitors,
that
it
just
spits
that
back
out
at
you,
I
tried
using
a
couple
free
services
and
all
spits
out
kindness
and
inevitable
kindness,
individual
tennis,.
C
A
E
A
E
So
well
we're
also
about
to
start
a
project
with
cyclica
trying
to
see
if
we
can
actually
look
at
predicting
tox
related
for
chemical
matter,
and
so
I
mean
we
could
also
put
one
of
these
or
a
couple
of
these
compounds
into
that
and
see.
If
that
turns
up
anything.
I
I
have
no
idea
at
this
point
how
successful
it
will
be,
but
we
could
certainly
try
and
incorporate
that
I
had
intended
to
look
at
the
nl
derivatives
that
we've
been
sending
right
right.
A
E
Yeah,
we
don't
have
any
kco2
data,
we
could
get
some,
but
the
other
thing
would
be
yeah
the
psychoactive
drug
screening
program.
I
was
looking
to
see
what
assets
they
have
available
and
I
mean
we've
tested
through
them
before.
I
know
it
does
say
that
you're
supposed
to
be
working
on
neurological
disorders
and
stuff,
but
they
they
do
test
other
things.
A
Because
we
don't
have
a
target,
yet
we
can't
do
anything
fancy
with
the
alpha
fold
two.
You
know
protein
dataset
thing
that
was
released
today
and
yesterday,
but
even
though
it's
not
actually
ready,
for
you
know,
analyzing
small
molecule
protein
binding,
it's
still
nice
to
play
around
with
the
structures,
but
this
project
is
not
ready
for
that.
Yet
so
nothing
we
can.
A
Maybe
we
can
do
it
for
pf-8p4
in
the
malaria
group.
Oh
somebody
did
that
overnight.
Yeah
used
this
to
predict
the
structure
of
that
of
that
protein
and
they
posted
it
yeah
within
hours.
It
was
interesting
but
yeah.
You
know
it's
not
intended
yet
for
small
molecule
protein
binding.
That's
a
the
next
level.
I
think.
A
Yeah
they
so
they've
done
the
predictions
for
for
the
human
proteome
and
for
other
species,
not
viruses,
but
for
other
species
and
deposited
them
with.
I
think.
H
A
A
H
H
A
I
think
so
yeah
yeah
yeah
definitely
for
sure,
and
it's
gonna
be
interesting
to
see
how
that
can
be
used
in
in
in
drug
discovery
projects.
I
guess
that's
the
next
phase,
but
it's
it's
fantastically
interesting
yeah.
I
mean
it's
yeah
beautiful,
to
see
these
things
for
sure
and
we'll
see
if
one
day
we
can
use
it
for
this
project,
but
yeah
making
I
mean
making
the
source
code
freely
available
for
the
for
the
calculation,
for
the
software
itself
is,
is
a
nice
step.
Yeah.