The Analysis of Gene Expression Data

by Parmigiani, Giovanni; Garrett, Elizabeth S.; Irizarry, Rafael A.; Zeger, Scott L.

• ISBN13: 9780387955773
• ISBN10: 0387955771
• Format: Hardcover
• Copyright: 2003-04-01
• Publisher: Springer Verlag

Summary

This book presents practical approaches for the analysis of data from gene expression microarrays. Each chapter describes the conceptual and methodological underpinning for a statistical tool and its implementation in software. Methods cover all aspects of statistical analysis of microarrays, from annotation and filtering to clustering and classification. Chapters are written by the developers of the software. All software packages described are free to academic users. The book includes coverage of various packages that are part of the Bioconductor project and several related R tools.The materials presented cover a range of software tools designed for varied audiences. Some chapters describe simple menu-driven software in a user-friendly fashion, and are designed to be accessible to microarray data analysts without formal quantitative training. Most chapters are directed at microarray data analysts with master-level training in computer science, biostatistics or bioinformatics. A minority of more advanced chapters are intended for doctoral students and researchers.The team of editors is from the Johns Hopkins Schools of Medicine and Public Health and has been involved with developing methods and software for microarray data analysis since the inception of this technology. Giovanni Parmigiani is Associate Professor of Oncology, Pathology and Biostatistics. He is the author of the book on "Modeling in Medical decision Making," a fellow of the ASA, and a recipient of theSavage Awards for Bayesian statistics. Elizabeth S. Garrett is Assistant Professor of Oncology and Biostatistics, and recipient of the Abbey Award for statistical education. Rafael A Irizarry is Assistant Professor of Biostatistics, and recipient of the Noether Award for non-parametric statistics. Scott L. Zeger is Professor and chair ofBiostatistics. He is co-author of the book "Longitudinal Data Analysis,"a fellow of the ASA and recipient of the Spiegelman Award for public health statistics.

Table of Contents

Preface	v
Contributors	xvii
The Analysis of Gene Expression Data: An Overview of Methods and Software	1	(45)
Giovanni Parmigiani
Elizabeth S. Garrett
Rafael A. Irizarry
Scott L. Zeger
Measuring Gene Expression Using Microarrays	1	(6)
Microarray Technologies	1	(3)
Sources of Variation in Gene Expression Measurements Using Microarrays	4	(1)
Phases of Microarray Data Analysis	5	(2)
Design of Microarray Experiments	7	(2)
Replication and Sample Size Considerations	7	(2)
Design of Two-Channel Arrays	9	(1)
Data Storage	9	(3)
Databases	9	(1)
Standards	10	(1)
Statistical Analysis Languages	11	(1)
Preprocessing	12	(4)
Image Analysis	12	(1)
Visualizations for Quality Control	12	(1)
Background Subtraction	13	(1)
Probe-level Analysis of Oligonucleotide Arrays	14	(1)
Within-Array Normalization of cDNA Arrays	15	(1)
Normalization Across Arrays	15	(1)
Screening for Differentially Expressed Genes	16	(3)
Estimation or Selection?	16	(1)
One Problem or Many?	17	(1)
Selection and False Discovery Rates	18	(1)
Beyond Two Groups	19	(1)
Challenges of Genome Biometry Analyses	19	(2)
Visualization and Unsupervised Analyses	21	(8)
Profile Visualization	21	(1)
Why Clustering?	22	(1)
Hierarchical Clustering	23	(2)
k-Means Clustering and Self-Organizing Maps	25	(1)
Model-Based Clustering	26	(1)
Principal Components Analysis	26	(1)
Multidimensional Scaling	27	(1)
Identifying Novel Molecular Subclasses	27	(1)
Time Series Analysis	28	(1)
Prediction	29	(4)
Prediction Tools	29	(1)
Dimension Reduction	30	(1)
Evaluation of Classifiers	30	(1)
Regression-Based Approaches	31	(1)
Classification Trees	31	(1)
Probabilistic Model-Based Classification	32	(1)
Discriminant Analysis	33	(1)
Nearest-Neighbor Classifiers	33	(1)
Support Vector Machines	33	(1)
Free and Open-Source Software	33	(3)
Whitehead Institute Tools	34	(1)
Eisen Lab Tools	34	(1)
TIGR Tools	34	(1)
GeneX and CyberT	35	(1)
Projects at NCBI	35	(1)
BRB	35	(1)
The OOML library	36	(1)
MatArray	36	(1)
BASE	36	(1)
Conclusion	36	(10)
Visualization and Annotation of Genomic Experiments	46	(27)
Robert Gentleman
Vincent Carey
Introduction	46	(1)
Motivations for Component-Based Software	47	(2)
Formalism	49	(1)
Bioconductor Software for Filtering, Exploring, and Interpreting Microarray Experiments	50	(8)
Formal Data Structures and Methods for Multiple Microarrays	50	(4)
Tools for Filtering Gene Expression Data: The Closure Concept	54	(1)
Expression Density Diagnostics: High-Throughput Exploratory Data Analysis for Microarrays	55	(2)
Annotation	57	(1)
Visualization	58	(6)
Chromosomes	59	(5)
Applications	64	(6)
A Case Study of Gene Filtering	64	(3)
Application of Expression Density Diagnostics	67	(3)
Conclusions	70	(3)
Bioconductor R Packages for Exploratory Analysis and Normalization of cDNA Microarray Data	73	(29)
Sandrine Dudoit
Jean Yee Hwa Yang
Introduction	73	(3)
Overview of Packages	73	(2)
Two-Color cDNA Microarray Experiments	75	(1)
Methods	76	(4)
Standards for Microarray Data	76	(1)
Object-Oriented Programming: Microarray Classes and Methods	77	(1)
Diagnostic Plots	78	(1)
Normalization Using Robust Local Regression	79	(1)
Application: Swirl Microarray Experiment	80	(1)
Software	81	(18)
Package marrayClasses---Classes and Methods for cDNA Microarray Data	81	(8)
Package marrayInput---Data Input for cDNA Microarrays	89	(2)
Package marrayPlots---Diagnostic Plots for cDNA Microarray Data	91	(5)
Package marrayNorm---Location and Scale Normalization for cDNA Microarray Data	96	(3)
Discussion	99	(3)
An R Package for Analyses of Affymetrix Oligonucleotide Arrays	102	(18)
Rafael A. Irizarry
Laurent Gautier
Leslie M. Cope
Introduction	102	(1)
Methods	103	(10)
Notation	103	(1)
The CEL/CDF Convention	104	(2)
Probe Pair Sets	106	(1)
Probe-Level Objects	107	(1)
Normalization	108	(3)
Exploratory Data Analysis of Probe--Level Data	111	(2)
Application	113	(2)
Expression Measures	113	(2)
Software	115	(3)
A Case Study	115	(3)
Extending the Package	118	(1)
Conclusion	118	(2)
DNA-Chip Analyzer (dChip)	120	(22)
Cheng Li
Wing Hung Wong
Introduction	120	(1)
Methods	121	(4)
Normalization of Arrays Based on an ``Invariant Set''	121	(1)
Model-Based Analysis of Oligonucleotide Arrays	122	(1)
Confidence Interval for Fold Change	122	(2)
Pooling Replicate Arrays Considering Measurement Accuracy	124	(1)
Software and Applications	125	(14)
Reading in Array Data Files	125	(2)
Viewing an Array Image	127	(2)
Normalizing Arrays	129	(1)
Viewing PM/MM Data	129	(2)
Calculating Model-Based Expression Indexes	131	(1)
Filter Genes	132	(1)
Hierarchical Clustering	133	(2)
Comparing Samples	135	(2)
Mapping Genes to Chromosomes	137	(1)
Sample Classification by Linear Discriminant Analysis	138	(1)
Discussion	139	(3)
Expression Profiler	142	(21)
Jaak Vilo
Misha Kapushesky
Patrick Kemmeren
Ugis Sarkans
Alvis Brazma
Introduction	142	(1)
EPCLUST	143	(8)
EPCLUST: Data Import	143	(1)
EPCLUST: Data Filtering	144	(2)
EPCLUST: Data Annotation	146	(1)
EPCLUST: Data Environment	147	(1)
EPCLUST: Data Analysis	148	(3)
URLMAP: Cross-Linking of the Analysis Results Between the Tools and Databases	151	(1)
EP:GO GeneOntology Browser	152	(1)
EP:PPI: Comparison of Protein Pairs and Expression	153	(1)
Pattern Discovery, Pattern Matching, and Visualization Tools	154	(1)
An Example of the Data Analysis and Visualizations Performed by the Tools in Expression Profiler	154	(5)
Integration of Expression Profiler with Public Microarray Databases	159	(1)
Conclusions	160	(3)
An S-PLUS Library for the Analysis and Visualization of Differential Expression	163	(22)
Jae K. Lee
Michael O'Connell
Introduction	163	(1)
Assessment of Differential Expression	164	(10)
Local Pooled Error	165	(4)
Tests for Differential Expression	169	(2)
Cluster Analysis and Visualization	171	(3)
Analysis of Melanoma Expression	174	(7)
Tests for Differential Expression	175	(3)
Cluster Analysis and Visualization	178	(2)
Annotation	180	(1)
Discussion	181	(4)
Dragon and Dragon View: Methods for the Annotation, Analysis, and Visualization of Large-Scale Gene Expression Data	185	(25)
Christopher M.L.S. Bouton
George Henry
Carlo Colantuoni
Jonathan Pevsner
Introduction	185	(4)
System and Methods	189	(10)
Overview of DRAGON	189	(1)
DRAGON's Hardware, Software, and Database Architecture	190	(2)
Cross-Referencing Information in DRAGON	192	(1)
The DRAGON Search and Annotate Tools	193	(3)
The DRAGON View Data Visualization Tools	196	(2)
DRAGON Gram: A Novel Visualization Tool	198	(1)
Implementation	199	(5)
Discussion and Conclusion	204	(6)
SNOMAD: Biologist-Friendly Web Tools for the Standardization and Normalization of Microarray Data	210	(19)
Carlo Colantuoni
George Henry
Christopher M.L.S. Bouton
Scott L. Zeger
Jonathan Pevsner
Introduction	210	(2)
Methods and Application	212	(13)
Overview of Experimental and Data Analysis Procedures	212	(2)
Background Subtraction	214	(1)
Global Mean Normalization	214	(1)
Standard Data Transformation and Visualization Methods	215	(2)
Local Mean Normalization Across Element Signal Intensity	217	(2)
Local Variance Correction Across Element Signal Intensity	219	(4)
Local Mean Normalization Across the Microarray Surface	223	(2)
Software	225	(1)
Discussion	226	(3)
Microarray Analysis Using the MicroArray Explorer	229	(25)
Peter F. Lemkin
Gregory C. Thornwall
Jai Evans
Introduction	229	(3)
Need for the Methodology	230	(1)
Basic Ideas Behind the Approach	231	(1)
Methods---Statistical and Informatics Basis	232	(7)
Analysis Paradigm	235	(3)
Particular Analysis Methods	238	(1)
Data Conversion	238	(1)
Software	239	(10)
System Design---Software Implementation	244	(3)
How to Download the Software	247	(1)
Strengths and Weaknesses of the Approach	248	(1)
Applications	249	(2)
Discussion	251	(3)
Parametric Empirical Bayes Methods for Microarrays	254	(18)
Michael A. Newton
Christina Kendziorski
Introduction	254	(2)
EB Methods	256	(5)
Canonical EB Example	256	(1)
General Model Structure: Two Conditions	256	(2)
Multiple Conditions	258	(1)
The Gamma--Gamma and Lognormal--Normal Models	259	(1)
Model Fitting	260	(1)
Software	261	(2)
Application	263	(6)
Discussion	269	(3)
SAM Thresholding and False Discovery Rates for Detecting Differential Gene Expression in DNA Microarrays	272	(19)
John D. Storey
Robert Tibshirani
Introduction	272	(1)
Methods and Applications	273	(10)
Multiple Hypothesis Testing	273	(2)
An Application	275	(1)
Forming the Test Statistics	276	(1)
Calculating the Null Distribution	277	(1)
The SAM Thresholding Procedure	278	(2)
Estimating False-Discovery Rates	280	(3)
Software	283	(6)
Obtaining the Software	283	(1)
Data Formats	283	(1)
Response Format	284	(1)
Example Input Data File for an Unpaired Problem	285	(1)
Block Permutations	285	(1)
Normalization of Experiments	285	(2)
Handling Missing Data	287	(1)
Running SAM	287	(1)
Format of the Significant Gene List	288	(1)
Discussion	289	(2)
Adaptive Gene Picking with Microarray Data: Detecting Important Low Abundance Signals	291	(22)
Yi Lin
Samuel T. Nadler
Hong Lan
Alan D. Attie
Brian S. Yandell
Introduction	291	(1)
Methods	292	(12)
Background Subtraction	292	(1)
Transformation to Approximate Normality	293	(2)
Differential Expression Across Conditions	295	(2)
Robust Center and Spread	297	(2)
Formal Evaluation of Significant Differential Expression	299	(2)
Simulation Studies	301	(3)
Comparison of Methods with E. coli Data	304	(1)
Software	304	(2)
Application	306	(7)
Diabetes and Obesity Studies	306	(2)
Software Example	308	(5)
MAANOVA: A Software Package for the Analysis of Spotted cDNA Microarray Experiments	313	(29)
Hao Wu
M. Kathleen Kerr
Xiangqin Cui
Gary A. Churchill
Introduction	313	(1)
Methods	314	(14)
Data Acquisition	315	(1)
ANOVA Models for Microarray Data	315	(2)
Experimental Design for Microarrays	317	(4)
Data Transformations	321	(1)
Algorithms for Computing ANOVA Estimates	322	(1)
Statistical Inference	323	(4)
Cluster Analysis	327	(1)
Software	328	(6)
Availability	328	(1)
Functionality	329	(5)
Data Analysis with MAANOVA	334	(5)
Discussion	339	(3)
GeneClust	342	(20)
Kim-Anh Do
Bradley Broom
Sijin Wen
Introduction	342	(1)
Methods	343	(4)
Algorithm	343	(1)
Choice of Cluster Size via the Gap Statistic	344	(2)
Supervised Gene Shaving for Class Discrimination	346	(1)
Software	347	(7)
The GeneShaving Package	347	(5)
GeneClust: A Faster Implementation of Gene Shaving	352	(2)
Applications	354	(4)
The Alon Colon Dataset	354	(2)
The NCI60 Dataset	356	(2)
Discussion	358	(4)
POE: Statistical Methods for Qualitative Analysis of Gene Expression	362	(26)
Elizabeth S. Garrett
Giovanni Parmigiani
Introduction	362	(2)
Methodology	364	(7)
Mixture Model for Gene Expression	364	(2)
Useful Representations of the Results	366	(1)
Bayesian Hierarchical Model Formulation	367	(1)
Restrictions to Remove Ambiguity in the Case of Only Two Components	368	(1)
Mining for Subsets of Genes	368	(2)
Creating Molecular Profiles	370	(1)
R Software Extension: POE	371	(10)
An Example of Using POE on Simulated Data	371	(1)
Estimating Posterior Probability of Expression Using poe.fit	372	(2)
Visualization Tools	374	(3)
Gene-Mining Functions	377	(2)
Molecular Profiling Tool	379	(2)
Results of POE Applied to Lung Cancer Data	381	(3)
Discussion and Future Work	384	(4)
Bayesian Decomposition	388	(21)
Michael F. Ochs
Introduction	388	(2)
Role of Signaling and Metabolic Pathways	388	(1)
Gene Expression Microarrays	389	(1)
Methods	390	(6)
Matrix Decomposition	390	(1)
Markov Chain Monte Carlo	391	(1)
Bayesian Framework	392	(1)
The Prior Distribution	393	(2)
Summary Statistics	395	(1)
Software	396	(2)
Implementation	396	(1)
Files and Installation	396	(1)
Issues in the Application of Bayesian Decomposition	397	(1)
Application of Bayesian Decomposition to Yeast Cell Cycle Data	398	(5)
Preparation of the Data	398	(1)
Running the Program	399	(1)
Visualizing the Output	400	(2)
Interpretation	402	(1)
Advantages of Bayesian Decomposition	403	(1)
Discussion	403	(6)
Bayesian Clustering of Gene Expression Dynamics	409	(19)
Paola Sebastiani
Marco Ramoni
Isaac S. Kohane
Introduction	409	(2)
Methods	411	(6)
Modeling Time	412	(1)
Probabilistic Scoring Metric	413	(2)
Heuristic Search	415	(1)
Statistical Diagnostics	416	(1)
Software	417	(3)
Screen 0: Welcome Screen	417	(1)
Screen 1: Getting Started	418	(1)
Screen 2: Analysis	418	(1)
Screen 3: Cluster Model	419	(1)
Screen 4: Pack and Go!	419	(1)
Application	420	(4)
Analysis	420	(1)
Statistical Diagnostics	421	(1)
Understanding the Model	421	(3)
Conclusions	424	(4)
Relevance Networks: A First Step Toward Finding Genetic Regulatory Networks Within Microarray Data	428	(19)
Atul J. Butte
Isaac S. Kohane
Introduction	428	(3)
Advantages of Relevance Networks	429	(2)
Methodology	431	(6)
Formal Definition of Relevance Networks	431	(1)
Finding Regulatory Networks in Phenotypic Data	432	(2)
Using Entropy and Mutual Information to Evaluate Gene--Gene Associations	434	(3)
Applications	437	(3)
Finding Pharmacogenomic Regulatory Networks	437	(2)
Setting the Threshold	439	(1)
Software	440	(7)
Index	447