Microarray Gene Expression Data Analysis A

This guide covers aspects of designing microarray experiments and analysing the data generated, including information on some of the tools that are available from non-commercial sources. Concepts and principles underpinning gene expression analysis are emphasised and wherever possible, the mathematics has been simplified. The guide is intended for use by graduates and researchers in bioinformatics and the life sciences and is also suitable for statisticians who are interested in the approaches currently used to study gene expression. Microarrays are an automated way of carrying out thousands of experiments at once, and allows scientists to obtain huge amounts of information very quickly Short, concise text on this difficult topic area Clear illustrations throughout Written by well-known teachers in the subject Provides insight into how to analyse the data produced from microarrays

Helen Causton is an experimental biologist who carried out some of the early studies on genome-wide transcriptional regulation in yeast using microarrays. She is Head of the Clinical Sciences Microarray Centre at Imperial College, University of London.

John Quackenbush is a principal investigator at The Institute for Genomic Research (TIGR). His research interests include development of software for microarray data analysis, gene indices and comparative genomics.

Alvis Brazma is a computer scientist who has been involved in microarray data analysis since 1998. He heads microarray informatics at the European Bioinformatics Institute and is in charge of establishing a public repository for microarray data.

Preface

ix

Acknowledgements

xi

Introduction

The central dogma of molecular biology

3

(1)

What are microarrays and how do they work?

4

(1)

Gene function and drug discovery

5

(2)

Data generation, processing and analysis: an overview

7

(3)

Data management

10

(4)

References

12

(2)

Experimental design

Experimental objectives and features of microarray data

14

(2)

General principles of experimental design

16

(10)

Reducing the number of variables

17

(1)

Time courses vs. independent data points

18

(1)

Replicates and repeated measurements

18

(4)

Reference samples

22

(1)

Exogenous (`spiked-in') controls

23

(1)

Dual labelling/dye swapping

24

(1)

Validation of results

25

(1)

Choice and preparation of samples

26

(4)

Obtaining the appropriate sample

26

(1)

Strain background

26

(1)

Mutants

26

(1)

Reagents

27

(1)

Sample and sample composition

27

(1)

Small amounts of sample: RNA amplification and pooling

28

(1)

Preparation of the labelled extract

28

(1)

Assessing the quality of the labelled extract

29

(1)

Choice and design of arrays

30

(3)

Choice of array platform

30

(1)

Oligonucleotides vs. PCR products

31

(1)

Replicate, guide and control features

32

(1)

Cross-hybridisation

32

(1)

Hybridisation, scanning and quality control

33

(2)

Long-term considerations

35

(5)

Record keeping

35

(1)

Standardisation

36

(1)

References

37

(3)

Image processing, normalisation and data transformation

Introduction

40

(1)

Preliminary processing of the data

41

(10)

Image analysis

41

(1)

Measuring and reporting expression

42

(2)

Saturated pixels

44

(1)

The appropriate number of pixels

44

(2)

Estimating background

46

(1)

Reporting expression with Affymetrix GeneChips™

47

(1)

Expression ratios: the starting point for sample comparison

48

(1)

Transformations of the expression ratio

49

(2)

Situations where expression does not correlate with spot intensity

51

(1)

Normalisation

51

(9)

Total intensity normalisation

54

(1)

Mean log centring

55

(1)

Linear regression

56

(1)

Chen's ratio statistics

57

(1)

Lowess normalisation

57

(2)

Global vs. local normalisation

59

(1)

Data filtering

60

(6)

Filtering low intensity data

61

(1)

Setting floors and ceilings

61

(1)

Use of replicate data

62

(1)

Experimental design strategies

62

(1)

Replicate filtering

63

(1)

Averaging replicate data

64

(2)

Identification of differentially expressed genes

66

(5)

Intensity-dependent estimation of differential expression

66

(1)

Analysis of variance

67

(2)

References

69

(2)

Analysis of gene expression data matrices

Introduction

71

(4)

Gene expression data matrices: their features and representations

75

(23)

Gene expression matrices

75

(4)

Representation of expression data as vector space - sample space and gene space

79

(2)

Distance and similarity measures in expression space

81

(1)

Euclidean, Minkowski, Manhattan, angle and chord distances

82

(3)

Pearson correlation distance, adjusting the mean and variance, correlation matrices, and the relationship between Euclidean and correlation distances

85

(4)

Spearman's rank correlation

89

(1)

Distances in discretised space, and mutual information

90

(2)

Principal component analysis, eigen-vectors and eigen-genes

92

(2)

Dealing with missing values

94

(1)

Representation of gene expression data by graphs (networks)

95

(1)

Gene expression matrix annotation

95

(3)

Clustering

98

(15)

Types of clustering

99

(1)

Hierarchical agglomerative clustering

100

(3)

Hierarchical divisive clustering

103

(1)

Non-hierarchical clustering - K-means

104

(1)

Self-organising maps and trees

105

(1)

Relationship between clustering and PCA

106

(1)

`Gene shaving'

107

(1)

Clustering in discretised space

108

(1)

Graph-based clustering

108

(1)

Bayesian or model-based clustering and fuzzy clustering

109

(1)

Clustering genes and samples - applications of clustering

110

(2)

Cluster scoring and validation

112

(1)

Classification algorithms and class prediction

113

(8)

Definition of the problem

114

(1)

Linear discriminants

115

(2)

Support vector machines

117

(1)

K-nearest neighbour method

118

(1)

Neural networks, decision trees and applications of classification

119

(1)

Partially supervised analysis

120

(1)

Class discovery

120

(1)

Time series analysis

121

(3)

Visualisation

124

(2)

Downstream from expression profile analysis

126

(8)

Identification of regulatory signals

128

(2)

References

130

(4)

Appendix: Non-commercial software

134

(9)

A.1 Statistical analysis

134

(2)

A.2 Normalisation, clustering and classification

136

(3)

A.3 Visualisation

139

(1)

A.4 Multifunctional software

139

(4)

References

141

(2)

Glossary

143

(11)

Index

154

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

FREE SHIPPING

Amazon no longer offers textbook rentals. We do!

Microarray Gene Expression Data Analysis A Beginner's Guide

Summary

Author Biography

Table of Contents

Supplemental Materials