►
Description
Open project meeting for Open Source Antibiotics Series 1, the Mur Ligases.
Full Project: https://github.com/opensourceantibiotics/murligase
Relevant GitHub Issue: https://github.com/opensourceantibiotics/murligase/issues/76
On the call: Professor Matthew Todd, Dr Edwin Tse, Yuhang Wang, Kato Leonard (UCL), Professor Chris Dowson, Laura Diaz Saez, Adrian Lloyd (University of Warwick), Lizbe Koekemoer (Diamond/Oxford), Dr Lori Ferrins, Dr Joe Eyermann (Northeastern University), Dr Chris Swain (Cambridge MedChem Consulting), Dr Jan Abendroth (UCB Biosciences), Dr Bart Staker (SSGCID).
A
All
right
welcome
to
the
muralage
meeting
on
14th
of
june
2022.,
I'm
on
a
single
screen,
so
I'm
going
to
share
screen,
even
though
sometimes
it's
a
bit
of
a
disaster,
if
you
haven't,
got
double
screen,
but
let's
see
how
we
go.
A
Okay,
so
there's
a
there's,
an
issue
here,
number
76
which
hopefully
has
everything-
and
I
think
that
I
guess
there's
two
main
things
today:
there'll
be
lots
of
little
bits,
but
two
main
things.
One
is
that
I
think
we
may
have
some
data
incoming
from
the
screening
of
enamine
compounds
from
from
joe
and
co
in
warwick,
for
that's
been
screened
by
by
chris
and
adrian.
A
After
we've
talked
about
those
two
things
and
of
course
we
can
talk
about
anything
else,
but
I
mean
I
guess
my
intention
was
to
to
hand
over
to
either
well
anyone
in
the
warwick
team
to
talk
about
either
of
those
two
things
either.
Chris
adrian
talk
about
data
or
laura
to
talk
about.
You
know
soak
in.
A
In
maybe
in
the
team's
meeting
we
may
have
a
team's
zoom
problem,
because
the
invite
may
have
appeared
to
be
from
teams.
C
A
C
A
C
No,
it's
the
same
as
the
dislikes
that
we
I
could
share
the
slides
from
last
time
from.
A
And
trying
to
get
at
the
yeah
what
what
it
is,
because
I
mean
one
of
the
really
key
conclusions-
was
this
idea
about
including
uma
and
the
structures
right
to
make
them
easier
to
do
and
yeah.
So
things
like
that,
you
know
the
sort
of
take-homes
for
people
who
weren't
here.
C
So
I
can
do
a
quick
update
for
the
rest
of
the
people
that
we've
enjoyed
in
the
structural
meeting.
So
basically,
we
have
been
doing
the
mirror.
E
and
mu
d
from
e
coli.
We've
been
working
with
them
for
extra
efforts
and
the
structures
were
mainly
able
structures
with
no
ligand
spawns,
which
meant
we
get
open
structures
and
the
atp
sites
that
we
want
to
target
to
do
the
dual
inhibition
or
multiple
inhibition
is
not
completely
formed.
So
we
are
not
getting
a
massive
amount
of
heat.
C
So,
to
do
that,
I
run
out
different
experiments
to
do
soaking
in
the
in
sorry
to
do
cockroaches
in
the
presence
of
different
ligands,
and
then
we
could
soak
into
those
structures
with
ligands
and
that
produce
three
structures
that
we
can
use
for
mu
d
and
one
with
adp
one
with
ampcp
and
one
with
the
uma,
the
monopeptide
so
and
we
can
use
that
the
one
that
seems
to
be
more
promising
because
the
atp
site
is
completely
empty,
is
the
monopeptide
structure,
so
that
one
is
the
first
priority
to
follow
up
and
trying
to
soak
components
into
it
and
jan
also
made
progress
and
got
a
structure
with
a
monopeptide
as
well
in
pseudomonas
new
d
and
he's
working
to
try
to
get
extra
things
and
extra
monopeptide
structures
as
well.
C
C
A
D
C
A
Okay,
just
before
we
go
to
adrian
two
quick
updates
from
us,
while
I'm
thinking
about
it,
as
you
were
finishing
up,
purification
of
the
putative
dual
inhibitors
that
the
you
and
kto
have
been
working
on,
and
I
think
you're,
are
you
nearly
ready
to
ship
those
molecules
to
warwick.
E
Yes,
let
me
just
show
you
this.
E
Yeah,
so
these
top
14
compounds
should
all
be
ready,
so
hopefully
I'll
be
able
to
send
them
end
of
this
week,
beginning
of
next
week
and
then
yeah
there's
a
bunch
of
these
competition
compounds
which
we're
working
on
now.
I
don't
think
I'll
have
them
ready
by
the
time
for
this
dress
shipment,
but
probably
in
a
few
weeks,
maybe.
C
A
Yeah,
so
a
few
teams
have
suggested
molecules
already
and
I
think,
we're
kind
of
expecting,
maybe
one
or
two
more
teams
to
come
with
molecules,
provided
we
spend
any
money
we
need
to
spend
by
the
end
of
the
month.
A
We'll
be
okay,
because
that's
when
the
grant
is
is
over,
so
we
we're
incentivized
to
spend
by
then
and
then
we
will
try
and
make
as
many
of
these
things
or
buy
as
many
of
these
things
as
we
possibly
can
all
right,
great
thanks,
er
and
then
the
other
thing
was
so
you
hang
can't
join
us
today.
He's
got
a
conflicting
thing.
He
is
very,
very
close
to
having
the
the
the
original
az
compound.
A
You
know
he's
been
working
on
changing
the
oh
nh2
to
try
and
improve
its
ability
to
target
wild-type
bacteria,
so
the
the
primary
nh2
is
meant
to
be.
You
know
helping
to
accumulate
the
compound
in
bacteria
he's
very
close
to
making
that
we've
got
a
trace.
That
we
think
is
the
right
molecule
with
the
right
mass
and
it's
pure
and
he's
just
trying
to
get
an
nmr
of
it
right
now.
A
But
with
that
methodology
he
should
be
able
to
make
a
few
other
compounds
too,
which
which
hopefully
will
hit
mercy
in
vitro
and
hopefully
will
accumulate
better
than
than
the
original
isaac
compounds.
So
he's
very,
very
nearly
there.
A
All
right
great
so
adrian
chris
did
you
want
to
present
the
new
data
yeah.
F
D
You
used
it,
you
know
they
did
all
that
now,
the
3d
printing
right,
you
could.
C
C
A
B
Have
you
mailed
adrian.
B
Should
we
come
back
to
this,
I.
B
A
A
Yep,
so
a
bunch
of
these
things
were
all
done.
The
atomized
thing
I
have
I've
not
been
back
in
terms
of
wise.
More
recently,
I
guess
there
was
one
thing
there
right
about
trying
to
figure
out
whether
the
atomized
compounds
were
wait.
There
was
one
action
here.
C
A
C
C
From
what
I
could
gather
from
becca's
notes,
it
might
be
composed
not
interest
interfered
with
the
essay,
so
they
were
tested,
but
then
no
data
can
be
extracted
from
them.
That's
what
I
understood,
but
because
away
at
the
moment,
so
we
will
have
to
wait
for
a
confirmation.
C
A
Yeah,
just
while
we've
got
them,
we
we
should
try
and
you
know,
exhaust
all
the
all
the
experiments
that
we
we
can.
C
A
And
then
yeah
we've
mentioned
the
competition
and
we're
in
there
was
the
open
global
health
library.
Chris
did
you
have
any
progress
on
that.
B
I
need
to
chase
them
again:
they've
had
they've
had
well,
I've
been
them
being
warwick,
warwick
contracting,
it's
the
same
old,
same
old
thing:
it's
stuck
in
a
contracting
black
hole;
okay,
all
right,
yeah
yeah!
I
think
it
just
reinforces,
for
you
know,
ongoing
and
future
things
I'm
trying
to
get.
You
know
agreed
sets
of
templates
signed
up
by
all
partners
beforehand,
because
you
know
so
I
don't
know
every
single
I've
had
to
do
it
with
life
arc
compounds
as
well.
B
That's
now
all
being
contracted
and
signed
up
for
that
hours
and
hours
and
hours
of
work.
I
found
for
me
to
go
through
the
contracts
and
try
and
persuade
people
why
why
we
just
need
to
sign
it
off
ditto,
every
other
one
as
well,
so
everything
that
comes
through
yeah,
okay,
I've
got
adrian
slides
through
now,
so
we're.
A
Thing
before
we
go
on,
there
was
this
cc4
car
scheme
that
we'd
mentioned
previously.
I
know
joe
and
laurie
are
familiar
with
that
which
offers
hit
to
lead.
I,
I
guess
I
just
want
to
keep
that
on
people's
radar.
A
In
case
we
do
come
up
with
new
hits
in
the
in
the
near
term
that,
where
this
this
scheme
might
be
suitable,
I
don't
know
if,
if
joe
or
laurie
or
anybody
else
has
looked
into
that
in
some
detail
and
advise
on
whether
we
might
be
suitable
for
it
or
are
we
too
early.
D
So
definitely
not
too
early,
I
mean
I,
I
know
ann
aiken
from
nih
who's
kind
of
leading
up
part
of
that
effort
and
x
x-a-z,
oh
really
so,
yeah.
I
think
this,
I'm
not
sure
kind
of
you
know
where
they
are
with.
You
know,
proposals
and
progress
and
so
on,
but
I
think
exactly
these
kinds
of
hits
if
they
get
confirmed
and
we
have
structure,
I
think
it'd
be
perfect
opportunity.
D
So
I
think
once
you
know
we
hear
from
adrian,
and
maybe
you
know
if
we
get
some
with
or
without
structure,
are
we
obviously
one
structure,
but
I
think
this
is
something
we
should
be
thinking.
You
know
we
should
talk
about
soon
about
maybe
submitting
if,
in
addition
to
the
grant
proposals
and
so
on,
but
anyway.
A
B
Cool
I'll
have
a
sharing
screen
then,
and
adrian
can
direct
me.
H
Just
a
word
on
the
cc4
carb
thing:
if
you
don't
mind,
I
mean
yeah,
it
looks
like
it's
in
a
nyad
program
right,
so
we
might
be
able
to
assist
with
that
since
we're
in
nyad
program
as
well
we're
supposed
to
work
with
them.
H
B
H
H
I
mean
we're
separate,
but
we're
they're,
separate
funding
and
all
that,
but
we're
pretty
well
connected
with
the
nyad
program
officers.
They
want
us
to
interact
with
all
their
other
nyad
programs
to
leverage
their
own
work.
So
I
don't
know
I
when
you,
if
you're
interested
in
that,
I
don't
know
anything
about
it,
but
I'd
be
happy
to
see
what
I
can
help
with
cool
great
okay.
D
Bart
I'll
catch
up
with
you
offline
about
about
that
and
how
we
can
kind
of
tag
team
from
both
of
our
ends.
On
that.
B
F
All
right,
so
this
is
rama's
a
group
effort
over
the
past
couple
months.
Two
names
that
happened
to
me
before
christopher
holds
and
kieran
duff.
They
are
two
final
year.
Students
who've
been
working
extremely
hard
over
the
past
few
weeks
on
some
of
the
compounds
that
we
now
have
an
interest
in.
B
I'm
not,
why
did
that
crash.
B
F
So
essentially
we're
the
starting
point
of
all
of
this
was
a
screen
of
an
enemy
library
targeting
atp
binding
sites
in
streptococcus,
elegantai
murdi,
which
was
a
project
that
was
initiated
between
joe
and
wayne
becker,
and
this
returned
a
number
of
hits
which
are
actually
shown
on
this
histogram
of
a
remaining
activities.
F
B
A
You
open
it
in,
I
think
you're
in
a
browser,
so
you
know,
can
you
open
it
in
powerpoint
itself.
G
B
F
B
F
D
F
Okay,
so
the
starting
point
for
looking
at
murder
and
murray
was
to
establish
what
the
kinetics
with
respect
to
atp
were
and
in
both
cases
there
were
hyperbolic
relationships
with
the
atp
concentration
and
velocity,
and
they
gave
us
good
estimates
of
the
kinetic
constants
for
atp
for
km
and
kcat,
and
I
want
you
to
remember
the
fact
that
these
are
hyperbolic,
because
this
isn't
always
the
case,
and
this
becomes
quite
relevant
later
on
but
anyway.
Well,
we
with
the
kinetic
constants.
F
We
got,
we
elected
to
drop
the
atp
concentration
to
a
third
of
km,
and
then
we
screened
becca's
hits
against
the
alligator
enzyme
against
modi
and
murray
next
slide.
F
And
the
result
of
that
was
that
we
got
some
inhibitors
which
were
dual
targeting,
for
example,
l06
and
some,
which
were
about
70
percent
remaining
at
percentage
inhibition
threshold
which
are
fo9,
n15
and
mo2.
F
Not
the
the
assay
that
we
used,
which
is
an
amplex
threat
based
assay,
wasn't
compatible
with
all
the
molecules
that
codes
screened,
and
one
of
these
is
j06,
which
is
pictured,
which
at
high
concentrations,
millimolar
concentrations
required
us
to
use
methyl
glycine
as
an
assay
chromophore.
F
When
we
did
this,
it
turned
out
to
be
a
good
inhibitor
against
both
birdie
and
next
slide.
Please.
F
So,
by
the
end
of
the
last
meeting,
they'd
started
generating
ic50
data,
so
this
is
powerful
if
f09
complex
ao1,
with
an
ic50
of
about
130
micromolar
and
a
hyperbolic
response,
and
we
then
plowed
through
the
other,
hits
that
we
got
always
cognizant
that
we
needed
to
positive
control,
which
is
on
the
right
hand,
side,
which
is
a
standard
adpcp
inhibitor,
which
gives
a
nice
d50
of
25
micromolar.
F
Similarly,
we
started
looking
at
modules
like
l06
again,
this
was
somewhat
less
potent,
but
we
could
still
measure
mic
and
next
slide.
F
And
then
we
got
onto
molecules
like
j06,
so
j06
has
a
21
micromolar
ic50,
which
corresponds
to
the
16.6
micromolar
ki,
which
meant
it
was
actually
binding
more
tightly
than
the
actual
standard
atp
analog
that
we'll
be
using.
So
if
you
carry
on
to
the
next
slide,.
F
So
all
that's
been
with
murder,
thus
far
with
j06
with
mercy.
The
ic50
was
equivalent
again.
It
was
quite
tight.
It
wasn't
quite
as
tight
as
adbcp
this
time,
but
still
with
a
respectable
ic50
of
16
micromolar
and
suggesting
a
ki.
If
it
is
competitive
of
12.4
micromolar
now
yeah
can
we
go
to
the
next
slide.
Please,
and
this
basically
jf6
steel,
targeting
both
pseudomonas
mer
e
and
the
d
with
roughly
equivalent
potency
and
if
it
could
go
to
the
next
slide.
F
So
the
other
murhi
hits
that
we
looked
at
with
pseudomonas
origination
were
ao2
again.
This
is
a
very,
very
high
ic-50
and
a
very
high
ki
and
the
data
does
show
some
scatter
and
we
would
probably
want
to
revisit
that.
F
F
F
So
we
extended
what
we
were
doing
to
looking
at
merge
c,
with
the
same
set
of
compounds
and
again
we
had
to
characterize
the
kinetics
of
mercy
with
respect
to
atp
and
at
low
concentrations
below
half
millimolar.
The
system
behaves
as
a
perfectly
ordinary
michaelis-menten
saturation
hyperbolic
saturation
with
a
km
and
okay
cap.
If
you
increase
the
concentration
of
atp,
you
go
into
substrate
inhibition,
and
this
becomes
interesting
because
this
suggests
that
there's
an
additional
atp
binding
sites.
F
Pilots
are
combined
in
two
ways:
either
way,
there's
more
than
one
interaction
at
the
lower
end
below
500
micromolar,
because
you
could
fit
the
data
to
the
case
meant
an
equation
as
well
as
an
equation
that
describes
subject
inhibition
and
there
was
very
little
difference
between
the
two,
we
basically
elected
to
drop
the
concentration
of
atp
on
the
basis
of
the
km
of
the
lower
atp
data
to
60
micromolar,
and
then
we
screened
becca's
hits
with
at
that
concentration
of
atp
with
udp
murnak,
and
this
is
what
we
found
next
slide.
F
So
we
found
numerous
hits
against
mercy.
We
found
compounds
that
targeted
c
and
d,
so
f,
o
nine
c
and
e
m
o
two
c
and
d
and
e
l
o
six.
So
at
the
moment
we're
at
this
stage
it
actually
goes
more
thoroughly
characterizing.
Those
particular
heads
against
pseudomonas
enzyme.
F
We
basically
gave
them
the
e
coli
murdy
and
mer
e
and
asked
them
to
see
what
they
could
find
with
the
compounds
that
we'd
already
started
to
characterize
and
again
they
start
with
the
kinetics
and
e
coli
is
again
hyperbolically
dependent
on
atp,
as
is
e
colimer
d
again
below
500
micromolar,
but
above
those
that
concentration,
the
enzyme
again
goes
into
substrate
inhibition,
which
you
can
see
in
the
bottom
right
hand
graph
again,
I'm
going
to
be
coming
back
to
that
at
the
end.
F
But
what
it
did
mean
is
that
if
we
worked
at
low
atp
concentrations,
we
could
probably
the
substrate
inhibition
that
we
saw
there,
but
I
don't
want
you
to
forget
that,
because
I'll
be
coming
back
to
it
next
slide.
Please.
F
So
essentially,
these
are
the
ic50
data
for
e-coli
and
e-coli
d.
On
each
panel,
the
ic50
is
run
at
two
atp
concentrations,
100
micromolar
and
20
micromolar,
and
so
a
competitive
inhibitor.
You
would
expect
to
show
an
ic50
which
increased
with
the
atp
concentration
and
with
j06,
which
is,
on
the
right
hand,
side.
That
appears
to
be
true
for
e,
and
it
appears
to
be
true
for
merge
d.
F
So
the
conclusion,
the
strict
conclusion
is
that
the
binding
of
j06
to
both
e
and
d
is
mutually
exclusive,
with
the
binding
of
atp
and
the
binding
of
e
coli,
the
d
to
l06
is
similarly
mutually
exclusive.
F
With
e
colimer
e,
there
was
insufficient
difference
between
the
two
ic50s
to
really
make
it
a
diagnosis,
it
could
easily
be
non-competitive
inhibition,
but
on
first
pass
it
does
seem
that
the
general
idea
that
you've
generated
an
atp
directed
inhibitor
in
silico
appears
to
bear
out
in
vitro.
So
we're
quite
happy
with
that.
F
F
F
We
gave
e
colimer
e
a
molecule
ap4a,
which
essentially
is
a
molecule
of
adenosine,
followed
by
four
phosphates,
followed
by
adenosine,
so
you
can
think
about
it
as
adp
linked
to
adp
and
it
turned
out
to
be
an
inhibitor
of
e
car,
like
the
e,
with
a
with
an
ic50
of
1.1
millimolar,
the
kinetics
look
sigmoid,
which
may
well
be
consistent
with
the
fact
that
there
are
essentially
two
binding
interactions,
but
the
interesting
thing
is,
you
can
actually
commercially
buy
uridine
tetraphosphoadenosine,
so
we're
in
the
process
of
purchasing
that
to
look
at
the
impact
of
that
molecule
because
it
might
just
be
that
it
could
be
possible
to
generate
inhibitors
which
actually
cross
both
sides.
F
F
So
the
last
thing
that
we
got
them
to
do
was
we
got
them
to
basically
trial,
compound
j06
and
lo6
in
straightforward,
low
volume,
10
migrated,
mics
and
essentially
the
take
home
from
all
of
this
is
that
wild
type
e
coli,
wild
type,
strep
alligatori
and
any
strain
of
pseudomonas
reginosis
we
tried,
did
not
have
an
mic
below
2.5
millimolar.
However,
a
tulsi
minus
mutant
of
e
coli
does
have
an
mic
of
around
about
104
micrograms
per
ml,
which
is
not
outstanding.
F
F
One
thing
that
does
come
to
mind
in
all
of
this
is
what
the
compound
actually
has
to
do.
Not
only
has
it
got
to
get
through
two
membranes,
it
gets
into
the
cytoplasm
and
it
faces
an
atp
concentration
of
the
order
of
about
10
millimolar,
depending
upon
growth
conditions.
F
A
F
A
Happy
to
see,
what's
you
know
available,
but
that's
yeah,
that's
great!.
F
I
mean
we
did
have
sort
of
internal
discussions
about
what
would
happen
if
we
tried
some
of
these
things
with
an
inhibitor
f1
f4
atpase,
for
example,
stuff
like
this
to
try
and
swing
the
balance,
as
they
were.
That's
a
that's
a
possibility.
I
mean
the
other
thing
that
we,
of
course
have
to
do
is
see
whether
actually
accumulating
any
peptidoglycan
curses
in
their
cells
as
well,
which
would
give
us
a
really
strong
indication
that
we
were
actually
targeting
the
right
thing.
B
B
D-Cyclocerine
is
variably
bacteriostatic
or
bactericidal,
depending
on
the
concentration
you're
using
on
the
other
hand,
phosphomycin
at
the
front
end
is
bactericidal,
so
I
was
going
to
have
a
chat
to
laura
about
the
structures
and
if,
if
we
might
even
be
able
to
think
about
making
conjugates
you
know
could
is
is,
is
there
a
possibility
to
make
a
phosphomycin
mercy
inhibitor
conjugate,
because
phosphomycin
is
great
for
getting
in
you
know
in
in
in
permeabilizing,
all
sorts
of
cells
gets
all
over
the
place.
B
Abilities,
we,
you
know,
there's
a
resistance
mechanism
that
modifies
c1
and
that
takes
out
the
activity.
So
we
know
what
inactivates
it
yeah.
G
A
C
B
Oh
cheers:
man
adrian
did
you
want
to
say
something
about
about
why
multi-targeting
is
probably
essential,
given
the
concentration
of
atp.
B
B
F
Yeah
I
mean
so
if
we've
got
an
atp
directed
inhibitor,
that's
just
just
about
doing
it,
then
its
actual
impact,
as
an
antimicrobial
in
any
sense,
is
going
to
be
dependent
upon
the
physiological
state
of
cells
which
is
going
to
dictate
the
atp
concentrations
within
the
cells.
And
if
you
look
at
people's
various
estimates
of
that,
that
can
be
anywhere
between
1
to
10
millimole.
F
Now
the
thing
is
that
we
we've
targeted
strep
alligatori,
which
is
a
facultative
anaerobe
pseudomonas
origin,
also,
which
is
an
obligate
arrow
and
e
coli,
which
is
patented,
and
so,
depending
upon
the
growth
conditions.
It's
not
really
reasonable
to
expect
the
atp
concentration
themselves.
F
To
be
the
same,
so
we
were
thinking
that
one
way
to
ensure
that
we
might
actually
swing
the
balance
is
if
we
were
to
start
investigating
whether
atp
inhibitors,
which
are
actually
some
of
which
are
currently
targeting
mycobacterial
atp
synthases
and
are
actually
in
clinical
trials.
B
So
so
you
know
again
again
thinking
about
what
we
we've
done.
This
in
the
past
is
age
reminded
me
about
15
years
ago,
with
the
pneumococcal
muscle,
inhibitor
yeah
we
did
optican
is
an
atp
synthase,
inhibitor
related
to
quinine,
hydroquinine
and,
and
that
dropped
the
mic
back
fourfold,
didn't
it
against
this
inhibitor
yeah.
B
So
for
tb,
I
think
that's
it's
an
interesting
proposition,
given
that
the
development
that
already
exists
with
the
inhibitors
and
and
again
being
being
mindful,
probably
to
maintain
the
multi-targeting,
we're
probably
going
to
have
to
retain
quite
a
small
size
of
molecule
and
thinking
about
again
chemical
conjugation,
possibly
with
some
of
these
alternative
inhibitors
at
an
at
an
early
point.
D
So,
what's
the
difference
between
atp
concentrations
and
a
bacteria
and
atp
concentrations
in
human
cells,
I
mean,
as
you
know,
there
are
more
than
20
30
40
kinase
directed
atp
dependent,
competitive
oncology
drugs.
I
mean
this
was
the
argument
that
was
made
25
30
years
ago
that
you
can
never
have
a
kinase
inhibitor,
because
atp
concentration
was
too
high.
So
so,
what's
what's
different
about
these
bacterial
cells
in
terms
of
atp
concentration
and
human
cell
atp
concentration,
I
think
the
answer
is
absolutely.
F
Nothing,
the
point
really
is
is
that
it's
they
both
will
have
millimolar
concentrations
of
atp.
What
it
comes
down
to
is
the
ability
to
generate
selective
and
potent
molecules
and
the
the
more
tightly
and
specifically
binding.
Those
molecules
are
the
more
likely
you're
going
to
be
able
to
get
good
competition.
F
The
types
of
type
binding
inhibitor
that
have
sort
of
nanomolar
type
affinities
bind
so
tightly
that
the
likelihood
is
that,
even
with
mill
mile
the
concentrations
of
atp
swimming
around,
there
is
little
chance.
So
you
get
the
competition
that
would
actually
cause
the
inhibitor
to
fail.
F
The
one
thing
that
did
occur
to
me
about
looking
about
thinking
about
these
things
is
that
some
of
the
more
ligase
inhibitors
that
we've
looked
at
in
before
appeared
not
to
bind
within
the
atp
binding
site
appear
to
actually
bind
towards
the
edge
of
it,
the
non-competitive
in
a
non-competitive
manner.
F
It
does
seem
to
me
that,
although
atp
finding
site
targeting
is
nice
because
it's
a
conserved
motif
throughout
the
merly
cases,
if
the
atp
concentration
is
going
to
be
something
that's
a
significant
factor,
it
may
not
be
a
good
idea
to
forget
about
the
fact
that
those
molecules
are
there,
because
that
does
get
around
a
problem.
I
think
the
only
issue
there
is
how
conserved
that
binding
site
actually
is
that
alternative
binding
sites,
because
you're
dependent
upon
that
conservation
for
multi-targeting.
D
So,
just
as
a
reminder,
there's
a
number
of
slides
on
github
that
compare
your
ligase
versus
some
known
human
kinases
and
we've
there's
a
difference
in
how
atp
binds
between
demure,
ligases
and
human
kinases
and
our
designs
have
been
focused
on
that
difference.
Yeah.
So
if
you
also
would
look
at
the
astrazeneca
your
c
paper,
where
they
compared
their
az,
your
c
inhibitors
against
a
panel
of
human
kinases,
you
will
see
that
there
was
selectivity.
D
I
would
expect
that
you
can
get
selectivity
between
your
c
d,
ligase,
inhibitor,
atp,
competitive
inhibitors
versus
human
you're,
always
going
to
have
some
kind
of
issue
with
selectivity.
I
mean
this.
This
is
the
bane
of
so
much
of
the
human
kinase
oncology
information
work,
but
people
have
spent
a
lot
of
time
and
have
been
able
to
develop
selective
human
kinase
inhibitors,
and
so
I
I
think
this
is
still
the
best
strategy
in
terms
of
going
after
a
mural
ligase
as
an
antibacterial
target.
Is
to
go
after
the
atp
site.
F
Everything,
I've
just
said
is
not
a
reason
for
not
doing
that
at
all.
It's
more
it.
It's
just
that,
because
we're
we're
dealing
with
fragments,
some
of
which
are
actually
for
fragments,
quite
potent
you
know
with
with
ki's
of
double
digit
micromolar.
F
D
Absolutely
I
mean,
I
think,
that's
the
focus.
That's
going
to
have
to
be
is
driving
the
potency
down
again.
You
know.
If
you
look
at
the
daisy
mercy
work,
we
did.
I
can't
remember
exactly
the
initial
hits,
but
you
know
you
know
we
drove
down
to
those
compounds
were
driven
down
by
a
thousand
folding
potency
and
that's
you
know.
D
As
we,
you
know,
move
forward
and
we
get
structures,
then
that's
going
to
be.
The
whole
goal
is
to
drive
the
potency
down,
hopefully
using
some
structure
and
medicinal
chemistry
effort.
So
I
I
totally
agree
I
mean
I
think
you
know
looking
at.
I
mean
I
think
it's
you
know
nice
for
your
students
to
be
able
to.
You
know
the
experiments
you
showed
with
with
the
microbiology.
I
mean
that,
that's
that's
that's.
You
know
nice
work,
but
yeah.
D
Obviously
we
need
to
you
know,
drive
the
potency
before
we
would
expect
to
see
any
significant
mics
yeah
anyway.
That's
great,
I
mean
it's
good.
I
mean
it's
all.
It's
all
good,
all
good.
D
D
I
do
I
do
have
a
question
I
guess
I'll
I'll
put
out
there
I
mean
so.
I
know
you
know
becca
selected
out
about
what
it
was
around
20
compounds
out
of
the
original
your
d
screen,
but
there
were
more
hits
that
were
potent
in
that
screen,
and
I
was
just
wondering
adrian,
you
know,
where
kind
of
where's
the
status
of
actually
profiling,
some
of
those
more
potent
hits
there
were
less
than
35
remaining
activity.
D
F
So
at
the
moment
we're
working
with
a
cutoff
of
about
30
percent,
so
we've
with
regard
to
turning
around
the
entire
library
against
the
cd
enf,
which
is
something
that
we
want
to
do
at
the
moment.
We're
still
establishing
that
stopped.
I
say
that
we've
got
is
up
to
the
job.
The
continuous
I
say
that
we
do
use
is,
but
that
does
have
a
significant
impact
upon
the
resources
required
to
run
it,
and
if
that's,
what
we
have
to
do,
that's
what
we
have
to
do.
F
What
steps
that
we've
taken
to
cut
down
the
work,
because,
in
actual
fact,
most
of
the
work
is
here,
is
not
actually
in
the
lab
it's
what
you
do
with
the
excel
data
afterwards.
That
takes
the
time.
So
we've
got
another
one
of
lab
members.
Who's
got
a
bit
more
expertise
in
actually
data
reduction
than
we
have
to
actually
come
up
with
a
matlab
routine,
reducing
progress
curves
to
to
as
to
basically
a
set
of
statistic
set
of
numbers
which
we
can
far
more
readily
interpret.
F
So
that
is
going
to
help
a
lot.
So
at
the
moment,
joe,
it's
a
work
in
progress,
but
we
will
get
there.
D
D
I
just
didn't
want
to
miss
out,
because
again,
I
think
there
were
some
compounds
that
looked
in
that
the
original
nerdy
becca's
original
year
d,
assay
that
looked
quite
interesting
below
30
percent-
that
for
whatever
reason,
weren't
selected-
and
I
I
guess
I
was
trying
to
understand-
you
know
why
they
haven't
been
further
profiled
but,
like
you
said,
you'll
get
there.
Yeah
yeah.
E
D
You
know
at
some
point
we
ultimately,
the
the
rubber
hits
the
road
in
terms
of
chemistry,
of
what
you
know,
series
which,
which
compounds,
which
hits
are
the
most
attractive
in
terms
of
trying
to
drive
them
digital
chemistry
program.
And
so
I
mean
that's,
that's
the
other
factor.
It
ultimately
comes
down
to.
If
you
kind
of
look
at
a
set
of
hits,
you
know
where
do
you
want
to
go
in
terms
of
chemistry?
What's
what's
the
most
tractable
chemistry
to
kind
of
follow
up
on?
D
So
I
think,
having
as
much
of
a
full
set
of
interests
of
the
compounds
from
the
hits
to
kind
of
have
a
discussion
about
where
we
want
to
go
in
terms
of
chemistry,
efforts,
whether
that's
with
a
niad
or
a
grant
proposal,
or
you
know
in
maths
lab
wherever
it's
good
just
to
have
the
kind
of
the
full
set
in
hand.
That's
all.
B
Yeah
yeah,
we
need
extra
pairs
of
hands,
adrian's
done
nothing
but
neoliga's
work,
for
I
can't
remember
when
pitto
of
the
members
of
the
lab,
so
I've
got
two
two
two
more
project
students
starting
for
work
over
the
summer,
so
hopefully
they'll
be
as
good
as
the
previous
two
we've
just
had,
and
they
can
get
them
to
crank
through
things.
But
to
be
honest,
pausing
and
investing
the
time
and
effort
and
getting
the
stopped
assay
done,
would
progress
an
awful
lot
of
things
a
lot
faster
it
would
they.
B
F
I
C
G
I
can
make
this
really
quick.
So
after
the
last
meeting
laura
sent
me
a
uma
compound
and
we
set
up
trace,
got
crystals
for
both
arsenic
tobacco,
muradi.
G
Got
data
last
week,
curiously
enough,
the
acetinobacter
structure
was
in
the
april
conformation
and
didn't
contain
any
uma.
I
need
to
follow
up
on
whether
that
was
human
error,
that
if
the
uma
was
just
wasn't
added
by
mistake,
but
the
good
news
is
that
I
have
now
a
structure
and
share
this,
so
we've
got
1.6
on
stream
density
or
1.67
structure
of
pseudomonas
mud
bound
to
uma.
What
is
very
curious
here,
we
got
another
molecule
sitting
over
here.
G
G
G
D
G
D
So
this
is
pseudomonas
your
d
right.
D
Does
it
make
any
sense
to
for
if
we
sent
you
any
compa,
if
any
compounds
for
any
of
these
fragments
were
sent
to
you,
does
it
I
mean
given
your
timeline
is?
Is
there
is
our
time
for
you
to
do
any
work
with
student
loans
from
your
d
or
you're,
pretty
well
kind
of
tapped
out
at
this
point.
G
G
I've
been
requesting
the
last
bit
of
protein
from
children's,
because
that's
where
the
protein
was
made
and
that's
where
the
majority
of
the
compounds
are
majority
of
the
protein
is
stored,
but
their
stocks
were
really
low
too.
Okay,
I
don't
know
if,
if
there's
really
strong
interest,
I
can
ask
bart
if
he
can
push
another
expression
in
purification,
but
I
don't
know
how
strong
their
appetite
for
that
is
and
how
many
hours
they
have
left
they
can
spend
on
this,
but
these
proteins
crystallize
within
like
a
week.
So
once
I
have
protein
I
can.
H
So
so
john,
I
would
say
that
you
know
we're
kind
of
we're
kind
of
hurting
for
protein
purification
right
now
because
of
personnel
turnover,
but
we
can
grow.
We
can
do
our
upskills,
like
we've
got
a
lot
of
capacity
right
now,
uh-huh
uh-huh,
so
I
would
say
we
should
submit
upskills
to
logan,
okay,
yeah.
G
D
Wondering
you
know-
and
we
have
e
coli
and
your
d
data,
that
that.
D
Adrian
presented,
excuse
me
that
adrian
presented
I'm
just
wondering
again.
I
guess
my
bias
is
again
a
little
bit
around.
You
know
the
mirror
seam
you're
deep,
what
you
call
those
not
dual
inhibitors,
multi,
multi
inhibitors,
it's
the
right
terminology.
I
guess
so
it'd
be
interesting
if
you
know
a
couple
of
the
fragments
to
try
to
get
a
a
pseudomonas,
your
d
structure,
but
anyway
we
can
talk
about
offline.
I
know
it's.