Statistics in high- -content biology content biology Statistics in - - PowerPoint PPT Presentation

Statistics in high Statistics in high-

• content biology

content biology

Rebecca Walls Rebecca Walls

Advanced Science & Technology Laboratory Advanced Science & Technology Laboratory

2 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Outline Outline

• Introduction and aim of high-content biology
• Predicting liver toxicity in vivo

in vivo

• Distinguishing distinct modes of compound action

3 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Current issues facing the pharmaceutical Current issues facing the pharmaceutical industry industry

• All pharmaceutical companies face high attrition of

compounds through the discovery and development process

• Two key issues that face project progression are
• Safety and toxicity

Safety and toxicity

• Efficacy in disease process

Efficacy in disease process

• Need to know more about the mechanism of action

and toxicity of our compounds at an earlier stage in the discovery process

• More information enables front-loading of risk, early

go/no-go decisions and improvements in toxicological attrition

4 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

High High-

• content biological assays

content biological assays

• Attempt to use in vitro

in vitro cell models to mimic the complexity of an in vivo in vivo situation

• Advanced imaging techniques used to generate

large, complex datasets describing the response of a population of cells to a compound

• Aim is to build predictive models or ‘fingerprints’

from the multiparametric assay data for well- characterised compounds that elicit known responses

• Fingerprints applied to new drugs to predict

biological mechanism of action and its toxicity

5 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Cell culture Cell culture

Cells Media layer

• Cells are extracted from some source

tissue e.g. rat hepatocytes, tumour derived cell-lines

• Cells are plated into

multi-well plates, typically hundred or thousands of cells per well

• Each well is like test tube

where we can test a single prototype drug

• Cells grown in

the well can be labelled and imaged

6 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

HCB cellular profiling HCB cellular profiling

Nucleus Nucleus DNA content Size Shape Cell division Fragmentation Micronuclei ER/ ER/Golgi Golgi Protein trafficking Secretion Mitochondria Mitochondria Viability Mass Activity Cellular distibution Pre-Apoptotic indicators Cytoskeleton Cytoskeleton Tubulin Actin Fibre content Length Mitotic arrest Apoptosis Apoptosis Membrane markers Blebbing Necrosis Cell Morphology Cell Morphology Count Area Form Roundness Length/Breadth Perimeter

General imaging indicators General imaging indicators

7 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Statistical challenges Statistical challenges

• Information captured for each feature is a dynamic

response to the compound over an 8-point dose- range

• Datasets possess three-dimensional cube-like

structure

• Traditional multivariate

approaches are difficult to apply to this type of data directly

FEATURES FEATURES COMPOUNDS COMPOUNDS DOSES DOSES

8 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Case study 1: Predicting liver toxicity Case study 1: Predicting liver toxicity in in vivo vivo

• Drug-induced liver toxicity is one of the most common causes of drug non-

approval

• Early in vitro identification of compounds with hepatotoxic risk would allow their

de-selection early in the drug development process

In the animal In the lab

Cell Death (Necrosis) Fatty Liver (Steatosis) Phospholipidosis Cholestasis

9 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Predicting Predicting steatosis steatosis -

• data

data

• Primary rat hepatotcytes treated with 60

compound set at a range of doses, consisting of known steatotics and non-steatotics

• Bespoke algorithms designed to quantify

differences in localisation and morphology of lipid droplets in the cells

• Generates 32 different continuous measurements

per cell

• Averaged over cell population to give well-level

measurements for each compound and dose combination

• Use partial least squares modelling (stepwise)

with the steatotic annotation as a binary response

• 5000

• 5000

• 5000

10 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Polynomial model Polynomial model

• Fit cubic polynomial to dose-response data for each feature
• t-statistics for each term in cubic form a new set of variables
• Only a small number of variables required to generate

greatest predictivity

• After cross-validation, polynomial model is approximately

10% better than range model

• 2
• 1

1 2

• 0.06 -0.04 -0.02 0.00 0.02 0.04

0.06 x y

Proportion of edge fat Proportion of edge fat – – non non-

• steatotic

steatotic

• 2
• 1

1 2 0.00 0.05 0.10 x y

Proportion of edge fat Proportion of edge fat – – steatotic steatotic

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Specificity Sensitivity 50 variables 1 variable

11 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Advantages of model Advantages of model

• Based on predictive scores, compounds can be ranked in order of steatotic effect
• Bootstrapping, incorporating random x-resampling, used to generate 95% confidence

intervals for the predicted score

• High confidence, high steatotic effect compounds can be de-selected

1 4 7 11 15 19 23 27 31 35 39 43 47 51 55 59

• 1.0
• 0.5

0.0 0.5 1.0 1.5 Compounds Steatotic effect

12 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Case study 2: Identifying distinct modes Case study 2: Identifying distinct modes

• f compound action
• f compound action
• Morphology high content assay developed specifically to

examine microtubules and actin filaments as oncology targets –

• Describes how drugs influence entire complex cellular phenotype

(i.e. multiple targets)

• 102 compounds screened through the morphology assay
• Primary aims are
• Identify which compounds are active in the assay i.e. which

Identify which compounds are active in the assay i.e. which are ‘hits’? are ‘hits’?

• Differentiate compound hits that have distinct morphological

Differentiate compound hits that have distinct morphological effects effects

• Cluster hits together that have similar effects

Cluster hits together that have similar effects

• 138 features for each compound, tested over 8 doses
• 310 control wells

13 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Principal components analysis Principal components analysis

• PCA used in an attempt to reduce

dimension of dataset, yielding 6 principal components which explain close to 80%

• f variation
• Mahalanobis distance is powerful means
• f determining how similar an unknown

sample is to a known one

• Differs from Euclidean distance in that it

takes into account the covariance between variables

• The Mahalanobis distance from a group
• f values with mean μ=(μ1, μ2, …, μp)T

and covariance matrix Σ for multivariate vector x=(x1, x2, …, xp)T is defined as

) ( ) ( ) (

1

μ μ − Σ − =

− x

x x DM

14 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Using the Mahalanobis distance Using the Mahalanobis distance

Squared Mahalanobis distances

Non Non-

• hits

hits Hits Hits

• Working on the PCA scores on the 6 principal components,

the covariance matrix of the control cloud was calculated

• For each compound at every dose, the squared

Mahalanobis distance to the centre of mass was calculated and compared to a chi-squared distribution with 6 degrees of freedom at some pre-chosen significance level, α.

• An adjustment was made to control the false discovery rate
• A compound with a significant result at at least

at least one of the doses along its range was deemed to be an ‘active hit’.

15 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Distinguishing distinct phenotypes Distinguishing distinct phenotypes

Buffer Compound A Compound B Compound C Compound D Compound E Compound F F Compound G

• Homogeneous nuclei

Homogeneous nuclei and cell shape and cell shape

• Stabilised cell

Stabilised cell-

• cell junctions

cell junctions – – results in ‘clumpy’ cells results in ‘clumpy’ cells

• No single cells

No single cells

• Aneuploidy

Aneuploidy – – big nuclei big nuclei

• Increased cell size

Increased cell size

16 Rebecca Walls, Non-Clinical Statistics Conference 2008, Leuven

Acknowledgements Acknowledgements

• Discovery Statistics

Discovery Statistics

• Chris

Chris Harbron Harbron

• Advanced Science and Technology Laboratory

Advanced Science and Technology Laboratory

• Ed

Ed Ainscow Ainscow

• Neil

Neil Carragher Carragher

• Andy

Andy Hargreaves Hargreaves

• Mike Sullivan

Mike Sullivan

• Helen Garside

Helen Garside

• James Pilling

James Pilling

• Lisa Rice

Lisa Rice

• Tom

Tom Houslay Houslay

• Peter

Peter Caie Caie

• Alex

Alex Ingleston Ingleston-

• Orme

Orme