Automated Extraction of Actionable Knowledge from Large Scale in-vitro pharmacology data

MedChemica | Jan 2017
Automated Extraction of Actionable
Knowledge from Large Scale in-vitro
pharmacology data:
the importance of stereochemistry in
life science research
Dr Al Dossetter
MedChemica Limited
Sheffield Stereochemistry – January 2017

Overview
•  Stereochemistry in real Drugs
– Not just chirality but stereochemistry
matters
•  What is important in drug designing?
•  Effects in chirality in Key in-vitro
•  Why do drugs fail in the clinic and the
staggering rise in R&D cost?
– Better rules for Medicinal Chemistry
•  How do we reduce the costs?
– Mining Actionable knowledge

Chiral Drug Molecules
Atazanavir

Atorvasta+n
Esomeprazole
Escitalopram

Sertraline
Topiramate
Eze+mibe

Conﬁrm
tumour
supression
with
in-‐Vivo
Xenograph

model
in
Nude
mouse

Inspira:on
–
A
Modern
Drug

Black
Box
Screening

Natural
Products

Halichondrin
B

Marine
natural
Product

Phenotypic
screening
–

arrested
cancer
cell
growth
Eribulin

(Halaven)
Approved
Nov
2010

Metasta:c
Breast
Cancer

Inhibitor
of
microtubule
dynamics

Inspiring
Med
Chem

Inspiring
synthesis

Cancer
Res.
2001,
61(3),
1013

Addressing
EFFICACY

DU-‐145
tumour
cell
line
–
growth
inhibited

rod
sphere
disc
Commercial fragments
access this space
N
N
H
F
O
H H
More interesting
Fragment?
Fragment needs
to be useful!
Shape Diversity Analysis of Commercial Fragment Libraries
Represents 14000+ fragments, from 4 vendors
includes random selection of compounds from
Chemonaut db

Nutlin
example
–
MDM2
binder
that
disrupts
interac:on
with
P53

The
inspira:onal
drug
discovery
program
in
‘Protein
Protein
Interac:on’
world

Directed

Screening

Use
knowledge
to

Select
compounds

Nutlin-‐3
Ro-‐7112
–
18nM

Structure
based
design
played
a

key
part
in
compound

op:misa:on

Vu
et
al,
ACS
Med
Chem
Le_,
2013.

This
and
other
compounds
have

‘sparked
an
understanding’
that

(Fragment)
libraries
for
PPIs

need
to
be
diﬀerent

Morelli,
Roche
et
al
MedChemComm,

2013,
DOI:
10.1039/c3md00018d

Cathepsin
K
–
Di-‐methoxy
surprise
–
Man
and
Machine

pIC50 7.95
LogD 0.67
HLM <2.0
Solubility 280μM
DTM ~1.0 mg/kg UID
Potent
Too polar / Renal Cl
PDB
-‐
97%
of
structures

Crawford,
J.J.;
Dosse_er,
A.G
J
Med
Chem.
2012,
55,
8827.

Dosse_er,
A.
G.
Bioorg.
Med.
Chem.
2010,
4405

Lewis
et
al,
J
Comput
Aided
Mol
Des,
2009,
23,
97–103

pIC50 8.2
LogD 2.8
HLM <1.0
Solubility >1400μM
DTM 0.01 mg/kg UID
High F% / stability
maximised
Increase in LogP,
Properties improved
Solubility

ΔpIC50 - 0.1
ΔLogD +1.4
ΔpSol +1.2
ΔHLM + 0.25
No renal Cl
low F%
ΔpIC50 +0.1
ΔLogD - 0.7
ΔpSol ~0.0
ΔHLM - 0.25
High F%
rat/Dog
Electrosta:c
poten:al
minima
between
oxygens

Approx
like
N
from
5-‐het,
new
compound
can
not

form
a
quinoline

Incr.
selec:vity

ΔpIC50 +0.1
ΔLogD - 0.7
ΔpSol ~0.0
ΔHLM - 0.25
High F%
rat/Dog

liver
kidneys
bladder
Dissolve (SOLUBILITY)
Cross
Membranes
(PERMEABILITY)
Metabolism
(Human Liver Microsomal,
Cytochrome P450 oxidation
and Inhibition)
Avoid
Excretion
Oral Dosing of Drugs
BBB (Blood Brain Barrier)
Brain difficult
Target Tissue
Survive pH range 1.5-8
Absorption
Distribution
Metabolism
Excretion

Effect of Chriality on Properties Effecting Drug Design - 1
S
-‐
omeprazole
R
-‐
omeprazole

Find all the known chirally pure enantiomers PAIRS with measured
biological ADMET properties.
Can we see a biological difference? How does is compare to physical
properties like aqueous solubility?

If physical properties
drove ADMET then
enantiomeric pairs
should have
equivalent ADMET
properties:
Enantiomeric pairs reveal that key medicinal chemistry parameters
vary more than simple physical property based models can explain
Andrew G. Leach et al, Med. Chem. Commun., 2012,3, 528-540.
1x
diﬀerence
between

matched
ena:omeric
pairs

=
whole
molecule

proper:es
will
be
enough.

Some:mes
true
but
not
useful
enough……

Effect of Chriality on Properties Effecting Drug Design - 2

Company

Ticker

Number of drugs
approved

R&D Spending
Per Drug ($Mil)

Total R&D
Spending
1997-2011 ($Mil)

AstraZeneca

AZN

5

11,790.93

58,955

GlaxoSmithKline

GSK

10

8,170.81

81,708

Sanofi

SNY

8

7,909.26

63,274

Pfizer Inc.

PFE

14

7,727.03

108,178

Roche Holding AG

RHHBY

11

7,803.77

85,841

Johnson & Johnson

JNJ

15

5,885.65

88,285

Eli Lilly & Co.

LLY

11

4,577.04

50,347

Abbott Laboratories

ABT

8

4,496.21

35,970

Merck & Co Inc

MRK

16

4,209.99

67,360

Bristol-Myers
Squibb Co.

BMY

11

4,152.26

45,675

Novartis AG

NVS

21

3,983.13

83,646

Amgen Inc.

AMGN

9

3,692.14

33,229

Sources: InnoThink Center For Research In Biomedical Innovation;
Thomson Reuters Fundamentals via FactSet Research Systems
The Truly Staggering Cost Of Inventing New Drugs
Matthew Herper - Forbes

Drug failures later in development are mainly due to EFFICACY and SAFETY

Actual spending / Chemistry everywhere
Paul, S. M. et al How to improve R&D productivity: the pharmaceutical
industry’s grand challenge, Nat. Rev. Drug Discovery 2010, 9, 203
Snap-Shot of a medium sized
companies R&D spend in one
year - $1.7 billion
For a period large pharma set targets at each stage of the process
– this was an “attrition model”
– this was unsuccessful and very wasteful
Better chemistry
Reduce the
number of
projects
Chemistry influences the
success and speed

What Causes Attrition in Development?
PK
7%
Lack
of

efficacy
in

man
46%
Adverse

effects
in
man
17%
Animal
toxicity
16%
Commercial

reasons
7%
Miscellaneous
7%
• Many
compounds
fail
in
development
through
inadequate

pharmacokineCcs
/
bioavailability
and
unacceptable
toxicological

profiles
in
addi:on
to
lack
of
efficacy
in
man

Big Data - Knowledge Based Design
The Life Science industry has woken
up to Big Data
•  Human Genome
•  Biological systems
•  Kinome
•  Metabolomics
•  Proteomics
•  3D structural information (CDC /
Protein Data Bank)
•  Literature and Patents (GVK Bio,
ChEMBL, Pubmed, PubChem)
•  Reaction informatics – what works,
what doesn’t
•  Document management
•  Regulatory submissions
Huge Opportunity in this area

What
about
research
data?

SAFE
DRUGS

‘Potency’

Do
not
sacriﬁce

The
be_er
it
is

the
lower
the
dose

Improved
tes+ng

in-‐vivo

with
fewer
animals

Clinical
linkage

to
protein
target

Can
test
In-‐Vivo

An:
SAR

e.g.
hERG,
Nav1.5,
5-‐HT2a…

Analysis
of
In-‐Vivo
data

Pﬁzer
–
rat
data

<0.2mg/Kg

Dose

Metabolism
&

Pharmacokine+cs

Be_er
design
so

dose
is
lower

Grand Rule
Database
Hughes
et
al,
Bioorg
Med
Chem

LeK.
2008,
18(17),
4872

Key
ﬁndings:

•  Stereochemistry is important in Drug hunting
•  There is a strong need powerful rules to understand med
chem better and reduce compound numbers and costs
How?
•  Secure sharing of large scale ADMET knowledge between
large Pharma is possible
•  The collaboration generated great synergy
•  Many findings are highly significant
•  Matched Molecular Pair Analysis (MMPA) is a great tool for
idea generation
•  The rules have been used in drug-discovery projects and
generated meaningful results
•  MMPA methodology can be extended to extract
pharmacophores

Fewer
compounds
designed
from
be_er
rules

from
data
analysis

•  Improved compounds quicker
•  Applicable ideas
•  Confident design decisions
•  Help when stuck
•  Clearly describable plans
•  Maximizing value from ADMET testing
•  Pursuing dead-end series
•  Pursuing dead-end projects
•  Running out of time or $
Essentials
Gains
Pains

Grand Rule database
Better medicinal chemistry by sharing knowledge not data & structures
MMP
finder
MCPairs=

Barriers
Broken
to
Sharing
Knowledge

Data
Integrity and
curation
Knowledge
extraction
algorithms
Consortium
building to
share
knowledge Into the minds of
chemists
✓

✓

✓

✓

Grand Rule
Database
MCPairs

MCPairs
Plarorm

•  Extract
rules
using
Advanced
Matched
Molecular
Pair
Analysis

•  Knowledge
is
captured
as
transforma:ons

–  divorced
from
structures
=>
sharable

Measured
Data
rule
finder Exploitable
Knowledge
MC Expert
Enumerator
System
Problem molecule
Solution molecules
Pharmacophores
& toxophores
SMARTS
matching
Alerts

Virtual
screening

Library
design

Protect
the
IP
jewels

MCPairs

•  Matched Molecular Pairs – Molecules
that differ only by a particular, well-
defined structural transformation
•  Transformation with environment capture –
MMPs can be recorded as transformations
from A B
•  Environment is essential to understand
chemistry
Statistical analysis
•  Learn what effect the transformation has had on ADMET properties in
the past
Griﬀen,
E.
et
al.
Matched
Molecular
Pairs
as
a
Medicinal
Chemistry
Tool.
Journal
of
Medicinal
Chemistry.
2011,
54(22),
pp.7739-‐7750.

Advanced
MMPA

Δ Data
A-B1
2
2
3
3
3
4
4
4
12
23
3
34
4
4
A

B

Magic
Methyl
–
Big
Potency
and
Property
improvements

Example
from
Leung,
C.S.;
Leung,
S.S.F.;
Tirado-‐Rives,
J.;
Jorgensen,
W.L.
J.
Med.
Chem.
2012,
55,
4489

Changing H to CH3 can bring big improvement even through this increases lipophilicity
Methyl group changes the shape of the molecule (often bringing ‘twists’ to rings)

Environment
really
ma_ers

HMe:

•  Median
Δlog(Solubility)

•  225
diﬀerent
environments

2.5log

1.5log

HMe:

•  Median Δlog(Clint)
Human microsomal
clearance
•  278 different
environments
We
can
see
in
the
context
the

shape
changes
that
bring
about

improved
proper:es

More
environment
=
right
detail

HMe Solubility:
•  225 different environments

HF
What
eﬀect
on
Clearance?

•  Median
Δlog(Clint)
Human
microsomal
clearance

•  37

diﬀerent
environments

2
fold
improvement
2
fold
worse

Increase

clearance

decrease

clearance

Rule
Example
1

Endpoint

mean±SD

count

LogD7.4

Solubility
–log(μM)

Cyp3A4
pIC50

- 0.880±0.542 n = 19
- 0.003±0.861 n = 14
- 0.111±0.431 n = 14

Rule Example 2
Endpoint mean±SD count
LogD7.4

Human
Liver
Microsomal
Clint

0.1±0.65 n = 14
- 0.39±0.12 n = 14

Rule Example 3
Endpoint mean±SD count
Human
Liver
Microsomal
Clint

Hepatocyte
Cells
Clint

- 0.35±0.25 n = 12
- 1.0 ±0.3 n = 9
MMPA can tell us occasions to make our molecules chiral and times not to….

Pharma 1 100k rules
Pharma 2 92k rules
Pharma 3 37k rules
5.8k rules in common (pre-merge) ~ 2%
New Rules 88k
~26% of total
Merge

Combining
data
yields
brand
new
rules

Gains:

300
-‐
900%

Merging knowledge – GRDv1

Early successes
From GRDv1 May 2014
31
J.
Med.
Chem.,
2015,
58
(23),
pp
9309–9333

DOI:
10.1021/acs.jmedchem.5b01312

- Fix hERG problem whilst maintaining potency
Waring et al, Med. Chem. Commun., (2011), 2, 775
Glucokinase Activators
MMPA
∆pEC50: -0.1 ∆logD: -0.6 ∆hERG pIC50 :-0.5
n=33 n=32 n=22
MMPA
∆pEC50: +0.3 ∆logD: +0.3 ∆hERG pIC50 :-0.3
n=20 n=23 n=19
MMPA
∆pEC50: -0.1 ∆logD: -0.6 ∆hERG pIC50 :-0.5
n=27 n=27 n=7

Knowledge Based Design – MPO
–  Novel more efficient core required, improve hERG for CD
–  CNS penetration, good potency and deliver tool for in vivo testing
McCoull, Dossetter et al, Med. Chem. Commun., (2013), 4, 456
ΔpIC50 -0.4
ΔlogD -1.8
ΔhERG pIC50 +0.4
Ghrelin Inverse agonists
MMPA
Cores
pIC50 9.9
logD 5.0
hERG pIC5 5.0
LLE 4.9
very potent
very lipophilic
ΔpIC50 +0.9
ΔlogD +0.2
ΔhERG pIC50 -0.3
pIC50 8.2
logD 1.3
hERG pIC50 4.4
LLE 6.9
ΔpIC50 -2.2
ΔlogD -2.2
ΔhERG pIC50 -0.7
100
compounds
made
LLE = lipophilic ligand efficiency:
LLE=pIC50-logD
LLE
6.4
LLE
6.9

A
Less
Simple
Example

Increase logD and gain solubility
Property
Number
of

Observa+ons

Direc+on
Mean
Change
Probability

logD
8
Increase
1.2
100%

Log(Solubility)
14
Increase
1.4
92%

What
is
the
eﬀect
on
lipophilicity
and

solubility?

Roche
data
is
inconclusive!
(2
pairs

for
logD,
1
pair
for
solubility)

logD
=
2.65

Kine:c
solubility
=
84
µg/ml

IC50
SST5
=
0.8
µM

logD
=
3.63

Kine:c
solubility
=
>452
µg/ml

IC50
SST5
=
0.19
µM

Ques+on:

Available

Sta+s+cs:

Roche

Example:

The application helped lead optimization in
project
•  193
compounds

•  Enumerated

Objective: improve metabolic stability
MMP
Enumeration
Calculated Property
Docking
8 compounds
synthesized

Solving
a
tBu
metabolism
issue

Benchmark

compound

Predicted
to
oﬀer
most
improvement
in
microsomal
stability
(in
at
least
1
species
/
assay)

R2

R1

tBu
Me
Et
iPr

99

392

16

64

78

410

53

550

99

288

78

515

41

35

98

327

92

372

24

247

35

128

24

62

60

395

39

445

3

21

20

27

57

89

54

89

•  Data shown are Clint for HLM and MLM (top and bottom, respectively)
R1
R2
R1
tBu

Roger Butlin
Rebecca Newton
Allan Jordan

…so
what
are
you

going
to
make

next…?

Comparison
of
Merck
in-‐house
MMPA
with
SALTMinerTM

Structure:
ADMET Issue: hERG
Lead A2A receptor antagonist
compound in Merck Parkinson's
project
138 suggestion molecules with
predicted improvement in hERG
binding
How many match the results of
Merck?
•  Also shows potent binding
to the hERG ion channel
•  Deng et al performed in-
house MMPA on hERG
binding compound data
and have published 18
resulting fluorobenzene
transformations, which they
have synthesized and
tested for hERG activity
Deng
et
al,

Bioorg.
&
Med
Chem
Let
(2015),

doi:
h_p://dx.doi.org/10.1016/
j.bmcl.2015.05.036

R
group:

Measured
hERG

pIC50
change

-‐1.187
-‐1.149
-‐1.038
-‐1.215
-‐1.157
-‐0.149
-‐1.487
-‐1.133

GRD
median

historic
pIC50

change

0
-‐0.171
-‐0.1
-‐0.283
-‐0.219
-‐0.318
-‐0.159
-‐0.103

Results:
8 out of the 18 fluorobenzene transformations produced by Merck were also suggested
by MCExpert to decrease hERG binding:
Searching the GRD for transformations that increase hERG there were none that
matched the remaining 10 of 18 transformations in the paper.
MCExpert also suggested an additional 50 fluorobenzene replacements to decrease
hERG binding NOT mentioned in the publication.

Fast building block access from CRO
collaboration
40
MCExpert
suggests
improved
building blocks
Specialist
synthesis CROs
access unique
chemistries
Rapid access to building
blocks that address
metabolism and solubility
issues
Mono & disubstituted
chiral piperidines
and pyrollidines
Chiral α methyl
aryl amines and
alcohols

Collaborators
and
Users
-‐
experience

Pharmacophores
and
Toxophores

by
extended
analysis
from
the
MMPA

PharmacophoresBigData Stats
Matched
Pairs
Finding
Public and in-
house potency
data

Mining
transform
sets
to
ﬁnd
inﬂuen:al
fragments

Identify the ‘Z’ fragments associated with a
significant number of potency increasing changes –
irrespective of what they are replaced with
‘Z’ is ‘worse than anything you replace it with’
Fragment A Fragment B

Change in binding
measurement
Public
Data
Find
Matched
Pairs
Find Potent
Fragments
+2.7

+3.2

+0.6

+0.6

Identify the ‘A’ fragments associated with a
significant number of potency decreasing changes
– irrespective of what they are replaced with
‘A’ is ‘better than anything you replace it with’
A

+2.1
+2.2

+1.4

+0.4

+1.8

Z

pKi/
pIC50
Compounds with
destructive fragment
Compounds with
constructive fragments
Generate
Pharmacophore
dyads
by

permuta:ng
all
the
fragments
with

the
shortest
path
between
them

Toxophores - Detailed, specific & transparent
Dopamine D2 receptor human
Actual: 9.5
Predicted: 9.1
Mean with: 8.0
Mean without: 6.6
Odds Ratio: 340
Dopamine Transporter
Actual: 9.1
Predicted: 8.6
Mean with: 8.3
Mean without: 6.7
Odds Ratio: 407
GABA-A
Actual: 9.0
Predicted: 8.7
Mean with: 8.0
Mean without: 6.8
Odds Ratio: 1506
β1 adrenergic receptor
Actual: 7.8
Predicted: 7.7
Mean with: 6.5
Mean without: 5.7
Odds Ratio: 1501
Find Potent
Fragments
Matched
Pairs
Finding
Find
Pharmacophore
Dyads
Public and in-
house potency
data

Prediction of unseen new molecules
The acid test…
•  Vascular endothelial growth factor receptor 2 tyrosine kinase (KDR)
•  Inhibitors have oncology and ophthalmic indications
•  Large dataset in CHEMBL
•  10 fold cross validated PLS model
•  Selected model by minimised RMSEP
46
Compounds 4466
Matched Pairs 288100
Fragments 678
Pharmacophore dyads 787
RMSEP 0.8
R2 0.64
Y-scrambled R2 0.0
ROC 0.95
Geomean odds ratio 80
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
● ●
●
●
●
●
●
●
●
●●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
● ●●
●●●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●
●●
●
●
●
●●●●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
4
6
8
10
5 7 9
pIC50_pred
pIC50

Novartis Predictions From Our Model
Domain of Applicability….
Actual: 8.4[1]
Predicted: 7.5
47
Actual: 7.6[1]
Predicted: 7.5
1. J MedChem(2016), Bold et al.
2. MedChem Lett (2016), Mainolfi et al.
Actual: 7.7[2]
Predicted: 7.1
Actual: 9.0[2]
Predicted: Out of Domain

Target
Number
of

compounds

Number
of

compound

pairs

Number
of

Fragments

Number
of

Pharmacophore

dyads
a|er

ﬁltering

R2
RMSEP
ROC

odds_ra:o

(geomean)

Acetylcholine esterase - human 383
27755
44
10
0.43
1.57
0.80
4

β 1 adrenergic receptor 505
145447
276
313
0.64
0.70
0.96
833

Androgen receptor 1064
113163
186
46
0.47
0.77
0.86
140

CB1 canabinnoid receptor 1104
88091
165
90
0.61
1.02
0.87
96

CB2 canabinnoid receptor 1112
82130
194
158
0.19
0.85
0.64
5.7

Dopamine D2 receptor - human 3873
230962
483
602
0.42
0.88
0.69
110

Dopamine D2 receptor - rat 1807
118736
267
377
0.29
0.85
0.78
125

Dopamine Transporter 1470
106969
282
336
0.58
0.73
0.88
141

GABA A receptor 848
39494
106
167
0.70
0.76
0.97
560

hERG ion channel 4189
242261
392
76
0.61
0.96
0.92
55

5HT2a receptor 642
50870
197
267
0.61
0.59
0.83
600

Monoamine oxidase 264
15439
44
11
0.12
1.25
0.48
181

Muscarinic acetylcholine receptor
M1
628
48200
97
510
0.62
0.94
0.89
48

µ opioid receptor 1128
37184
33
11
0.69
1.30
0.87
81

Critical safety target analysis
•  Build models using 10-fold cross validated PLS
•  Assess using ROC / BEDROC, R2 vs 100 fold y-scrambled R2 and geomean odds ratio
48
Public
Data
Find
Matche
d Pairs
Pharmacophores
Find
Pharmacophore
dyads
Find Potent
Fragments

MCBiophore GUI screenshot
Assay Image Mean_with Mean_without PLS_coeff Path SMARTS1 SMARTS2 n_examples odds_ratio
2
VEGFR 8.3 6.4 0.71 [c]c[c][c] Cc1ccc[c]c1[n] [c]/C(=N/O)/C 18 259.8
3
VEGFR 8.2 6.4 0.17 [c] [c]c1cc(cc[c]1)/C(=N/OC)/CCc1ccc[c]c1[n] 17 257.6
0
VEGFR 8.1 6.4 -0.01 [CH3] [cH]c([cH])/C(=N/OC)/C[c]/C(=N/OCC)/C 8 20.2
1
VEGFR 8.1 6.4 -0.01 [CH3] [c]c1cc(cc[c]1)/C(=N/OC)/C[c]/C(=N/OCC)/C 8 151.2
Detailed
results in
excel

Matched Molecular Pair
data A data B
data C data D
data E data F
Chemical Transformations
Δ data A B
Δ data C D
Δ data E F
Chemical Transformations
Δ data A B
Δ data C D
Δ data E F
Δ data G H
Δ data I J
Δ data K L
Matched Molecular Pair Analysis (MMPA) enables SAR sharing
Without sharing underlying structures and data
Grand
Rule
Database
Enumeration
Rate-My-Idea
GRD-Browser
ChEMBL Tox database Toxophores
MC-
Biophore
MCPairs

A Collaboration of the willing
Craig Bruce OE
John Cumming Roche
David Cosgrove C4XD
Andy Grant★
Martin Harrison Elixir
Huw Jones Base360
Al Rabow Consulting
David Riley AZ
Graeme Robb AZ
Attilla Ting AZ
Howard Tucker retired
Dan Warner Myjar
Steve St-Galley Syngenta
David Wood JDR
Lauren Reid MedChemica
Shane Monague MedChemica
Jessica Stacey MedChemica
Andy Barker Consulting
Pat Barton AZ
Andy Davis AZ
Andrew Griffin Elixir
Phil Jewsbury AZ
Mike Snowden AZ
Peter Sjo AZ
Martin Packer AZ
Manos Perros Entasis Therapeutics
Nick Tomkinson AZ
Martin Stahl Roche
Jerome Hert Roche
Martin Blapp Roche
Torsten Schindler Roche
Paula Petrone Roche
Christian Kramer Roche
Jeff Blaney Genentech
Hao Zheng Genentech
Slaton Lipscomb Genentech
Alberto Gobbi Genentech

Appendix

References on Lean in R&D
Sewing, A Drug Disco. Techno, 2009, DOI, 10.1016/j.ddtec,2008,12.002
Andersson S et al, Making medicinal chemistry more effective--application of Lean Sigma to improve processes, speed and quality.
Drug Discov Today. 2009 Jun;14(11-12):598-604.
Johnstone, C.; Pairaudeau, G.;Pettersson, J. A.; Creativity, innovation and lean sigma: a controversial combination? Drug Discov
Today. 2011 Jan;16(1-2):50-7
Robb, G.R.; McKerrecher, D.;Newcombe, N.J.;Waring, M.J. A chemistry wiki to facilitate and enhance compound design in drug
discovery. Drug Discov Today. 2013 Feb;18(3-4):141-7.
Plowright, A.T.; Johnstone, C.; Kihlberg, J.; Pettersson, J.; Robb, G.; Thompson, R.A.; Hypothesis driven drug design: improving quality
and effectiveness of the design-make-test-analyse cycle. Drug Discov Today. 2012 Jan;17(1-2):56-62
Baldwin, E.T., Metrics and the effective computational scientist: process, quality and communication. Drug Discov Today. 2012 Sep;
17(17-18):935-41.
Cumming, J.G.; Winter, J.P.; Poirrette, A. Better compounds faster: the development and exploitation of a desktop predictive chemistry
toolkit. Drug Discov Today. 2012 Sep;17(17-18):923-7.
Baede, E.J.; Bekker, E.J.W.; Cronin, D.;Integrated project views: decision support platform for drug discovery project teams. J Chem Inf
Model. 2012 Jun 25;52(6):1438-49.
Contrast to:-
MacDonald, J. F.; Smith, P. W. Lead Optimization in 12 months? True confession of a chemistry Team Drug Discovery Today, 2001, 6, 18,
947
•  Parallel Screening was an important outcome of the application of Lean
Manufacturing
•  Reducing the work in progress to avoid spreading chemistry effort was
important
•  The best results were achieved by encouraging team work and increasing
CLARITY through effective COMMUNICATION

Human
Element
-‐
Chemists
like
their
own
ideas…….

They
like
the
look
of
it

• 
Asked
19
chemists
to
look

through
a
set
of
fragments
and

choose
what
they
considered
the

‘best
ones’
to
follow
up

• 
When
asked
how
they
choose

them
they
self
report
that
it
was

mul:-‐factorial

• 
Analysis
shows
they
were

chosen
on
Ring
topology
and

Func:onal
groups
(not
really
on

size
or
lipophilicity)

Kutchukian,
P.S.
et
al
‘Inside
the
mind
of
the
Medicinal
Chemist’
PLOS
one
2012,
doi:
10.1371/journal.pone.0048476

See
also
Cheshire,
D.
R.
‘How
well
do
Medicinal
Chemists
learn
from
Experience,
Drug
Discov.
Today,
2011,
16,
(17/18),

817.
Leeson,
P.D.;
Springthorpe,
B.
The
inﬂuence
of
drug-‐like
concepts
on
decision-‐making
in
med.
Chem.

Nat.
Rev.
Drug
Discov.
2007,
6,
881.

But the literature says it’s lipophilicity
Does it?

‘The
focus
on
Ro5
is
oral

absorp:on
and
the
rule

neither
quan:ﬁes
the
risk
of

failure
associated
with
non-‐
compliance
nor
provides

guidance
as
to
how
sub-‐
op:mal
characteris:cs
of

compliant
compounds
might

be
improved’

Kenny,
P.
W.;
Montanari,
C.
A.

J.
Comput
Aided
Mol
Des,

2013,
27,
1-‐13.

See
also:

Carlson,
H.
A.
J.
Chem.Inf.Model,
2013,

dx.doi.org/10.1021/ci4004249

Automated Extraction of Actionable Knowledge from Large Scale in-vitro pharmacology data

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (7)

Similar to Automated Extraction of Actionable Knowledge from Large Scale in-vitro pharmacology data

Similar to Automated Extraction of Actionable Knowledge from Large Scale in-vitro pharmacology data (20)

Recently uploaded

Recently uploaded (20)

Automated Extraction of Actionable Knowledge from Large Scale in-vitro pharmacology data