Design and Analysis of DNA Microarray Investigations

by Korn, Edward Lee; Simon, Richard M.; McShane, Lisa M.; Radmacher, Michael D.; Wright, George W.

• ISBN13: 9780387001357
• ISBN10: 0387001352
• Format: Hardcover
• Copyright: 2004-02-01
• Publisher: Springer Verlag

Summary

This book is targeted to biologists with limited statistical background and to statisticians and computer scientists interested in being effective collaborators on multi-disciplinary DNA microarray projects. State-of-the-art analysis methods are presented with minimal mathematical notation and a focus on concepts. This book is unique because it is authored by statisticians at the National Cancer Institute who are actively involved in the application of microarray technology. Many laboratories are not equipped to effectively design and analyze studies that take advantage of the promise of microarrays. Many of the software packages available to biologists were developed without involvement of statisticians experienced in such studies and contain tools that may not be optimal for particular applications. This book provides a sound preparation for designing microarray studies that have clear objectives, and for selecting analysis tools and strategies that provide clear and valid answers. The book offers an in depth understanding of the design and analysis of experiments utilizing microarrays and should benefit scientists regardless of what software packages they prefer. In order to provide all readers with hands on experience in data analysis, it includes an Appendix tutorial on the use of BRB-ArrayTools and step by step analyses of several major datasets using this software which is freely available from the National Cancer Institute for non-commercial use. The authors are current or former members of the Biometric Research Branch at the National Cancer Institute. They have collaborated on major biomedical studies utilizing microarrays and in the development of statistical methodology for the design and analysis of microarray investigations. Dr. Simon, chief of the branch, is also the architect of BRB-ArrayTools.

Table of Contents

Acknowledgments	v
Introduction	1	(4)
DNA Microarray Technology	5	(6)
Overview	5	(1)
Measuring Label Intensity	5	(1)
Labeling Methods	6	(1)
Printed Microarrays	7	(2)
Affymetrix GeneChip™ Arrays	9	(1)
Other Microarray Platforms	10	(1)
Design of DNA Microarray Experiments	11	(18)
Introduction	11	(1)
Study Objectives	12	(1)
Class Comparison	12	(1)
Class Prediction	13	(1)
Class Discovery	13	(1)
Pathway Analysis	13	(1)
Comparing Two RNA Samples	13	(1)
Sources of Variation and Levels of Replication	14	(2)
Pooling of Samples	16	(1)
Pairing Samples on Dual-Label Microarrays	17	(4)
The Reference Design	17	(2)
The Balanced Block Design	19	(1)
The Loop Design	20	(1)
Reverse Labeling (Dye Swap)	21	(2)
Number of Biological Replicates Needed	23	(6)
Image Analysis	29	(10)
Image Generation	29	(1)
Image Analysis for cDNA Microarrays	30	(5)
Image Display	30	(1)
Gridding	30	(1)
Segmentation	31	(1)
Foreground Intensity Extraction	32	(1)
Background Correction	33	(1)
Image Output File	34	(1)
Image Analysis for Affymetrix GeneChip™	35	(4)
Quality Control	39	(14)
Introduction	39	(1)
Probe-Level Quality Control for Two-Color Arrays	40	(4)
Visual Inspection of the Image File	40	(1)
Spots Flagged at Image Analysis	40	(1)
Spot Size	41	(1)
Weak Signal	42	(1)
Large Relative Background Intensity	43	(1)
Gene Level Quality Control for Two-Color Arrays	44	(3)
Poor Hybridization and Printing	45	(1)
Probe Quality Control Based on Duplicate Spots	45	(1)
Low Variance Genes	46	(1)
Array-Level Quality Control for Two-Color Arrays	47	(1)
Quality Control for GeneChip™ Arrays	48	(2)
Data Imputation	50	(3)
Array Normalization	53	(12)
Introduction	53	(1)
Choice of Genes for Normalization	53	(2)
Biologically Defined Housekeeping Genes	53	(1)
Spiked Controls	54	(1)
Normalize Using All Genes	55	(1)
Identification of Housekeeping Genes Based on Observed Data	55	(1)
Normalization Methods for Two-Color Arrays	55	(6)
Linear or Global Normalization	56	(1)
Intensity-Based Normalization	57	(2)
Location-Based Normalization	59	(2)
Combination Location and Intensity Normalization	61	(1)
Normalization of GeneChip™ Arrays	61	(4)
Linear or Global Normalization	61	(1)
Intensity-Based Normalization	62	(3)
Class Comparison	65	(30)
Introduction	65	(1)
Examining Whether a Single Gene is Differentially Expressed Between Classes	66	(9)
t-Test	67	(1)
Permutation Tests	68	(3)
More Than Two Classes	71	(2)
Paired-Specimen Data	73	(2)
Identifying Which Genes Are Differentially Expressed Between Classes	75	(9)
Controlling for No False Positives	76	(4)
Controlling the Number of False Positives	80	(1)
Controlling the False Discovery Proportion	81	(3)
Experiments with Very Few Specimens from Each Class	84	(2)
Global Tests of Gene Expression Differences Between Classes	86	(2)
Experiments with a Single Specimen from Each Class	88	(2)
Regression Model Analysis; Generalizations of Class Comparison	90	(1)
Evaluating Associations of Gene Expression to Survival	91	(1)
Models for Nonreference Designs on Dual-Label Arrays	92	(3)
Class Prediction	95	(26)
Introduction	95	(2)
Feature Selection	97	(1)
Class Prediction Methods	98	(10)
Nomenclature	98	(1)
Discriminant Analysis	98	(3)
Variants of Diagonal Linear Discriminant Analysis	101	(2)
Nearest Neighbor Classification	103	(1)
Classification Trees	104	(2)
Support Vector Machines	106	(1)
Comparison of Methods	107	(1)
Estimating the Error Rate of the Predictor	108	(6)
Bias of the Re-Substitution Estimate	108	(2)
Cross-Validation and Bootstrap Estimates of Error Rate	110	(2)
Reporting Error Rates	112	(1)
Statistical Significance of the Error Rate	113	(1)
Validation Dataset	113	(1)
Example	114	(4)
Prognostic Prediction	118	(3)
Class Discovery	121	(36)
Introduction	121	(1)
Similarity and Distance Metrics	122	(3)
Graphical Displays	125	(6)
Classical Multidimensional Scaling	125	(6)
Nonmetric Multidimensional Scaling	131	(1)
Clustering Algorithms	131	(15)
Hierarchical Clustering	131	(7)
k-Means Clustering	138	(4)
Self-Organizing Maps	142	(3)
Other Clustering Procedures	145	(1)
Assessing the Validity of Clusters	146	(11)
Global Tests of Clustering	148	(2)
Estimating the Number of Clusters	150	(2)
Assessing Reproduciblity of Individual Clusters	152	(5)
A Basic Biology of Gene Expression	157	(8)
Introduction	157	(8)
B Description of Gene Expression Datasets Used as Examples	165	(4)
Introduction	165	(1)
Bittner Melanoma Data	165	(1)
Luo Prostate Data	166	(1)
Perou Breast Data	166	(1)
Tamayo HL-60 Data	167	(1)
Hedenfalk Breast Cancer Data	168	(1)
C BRB-ArrayTools	169	(16)
Software Description	169	(2)
Analysis of Bittner Melanoma Data	171	(7)
Analysis of Perou Breast Cancer Chemotherapy Data	178	(4)
Analysis of Hedenfalk Breast Cancer Data	182	(3)
References	185	(10)
Index	195