I Preprocessing data from genomic experiments ................. 1
1 Preprocessing Overview ..................................... 3
W. Huber, R.A. Irizarry, and R. Gentleman
1.1 Introduction ............................................... 3
1.2 Tasks ...................................................... 4
1.2.1 Prerequisites ....................................... 5
1.2.2 Stepwise and integrated approaches .................. 5
1.3 Data structures ............................................ 6
1.3.1 Data sources ........................................ 6
1.3.2 Facilities in R and Bioconductor .................... 7
1.4 Statistical background ..................................... 8
1.4.1 An error model ...................................... 9
1.4.2 The variance-bias trade-off ........................ 11
1.4.3 Sensitivity and specificity of probes .............. 11
1.5 Conclusion ................................................ 12
2 Preprocessing High-density Oligonucleotide Arrays ......... 13
B.M. Bolstad, R.A. Irizarry, L. Gautier, and Z. Wu
2.1 Introduction .............................................. 13
2.2 Importing and accessing probe-level data .................. 15
2.2.1 Importing .......................................... 15
2.2.2 Examining probe-level data ......................... 15
2.3 Background adjustment and normalization ................... 18
2.3.1 Background adjustment .............................. 18
2.3.2 Normalization ...................................... 20
2.3.3 vsn ................................................ 24
2.4 Summarization ............................................. 25
2.4.1 expresso ........................................... 25
2.4.2 threestep .......................................... 26
2.4.3 RMA ................................................ 27
2.4.4 GCRMA .............................................. 27
2.4.5 affypdnn ........................................... 28
2.5 Assessing preprocessing methods ........................... 29
2.5.1 Carrying out the assessment ........................ 30
2.6 Conclusion ................................................ 32
3 Quality Assessment of Affymetrix GeneChip Data ............ 33
B.M. Bolstad, F. Collin, J. Brettschneider, К. Simpson,
L. Cope, R.A. Irizarry, and T.P. Speed
3.1 Introduction .............................................. 33
3.2 Exploratory data analysis ................................. 34
3.2.1 Multi-array approaches ............................ 35
3.3 Affymetrix quality assessment metrics ..................... 37
3.4 RNA degradation ........................................... 38
3.5 Probe level models ........................................ 41
3.5.1 Quality diagnostics using PLM ...................... 42
3.6 Conclusion ................................................ 47
4 Preprocessing Two-Color Spotted Arrays .................... 49
Y.H. Yang and A.C. Paquet
4.1 Introduction .............................................. 49
4.2 Two-color spotted microarrays ............................. 50
4.2.1 Illustrative data ................................. 50
4.3 Importing and accessing probe-level data .................. 51
4.3.1 Importing .......................................... 51
4.3.2 Reading target information ......................... 52
4.3.3 Reading probe-related information .................. 53
4.3.4 Reading probe and background intensities ........... 54
4.3.5 Data structure: the marrayRaw class ................ 54
4.3.6 Accessing the data ................................. 56
4.3.7 Subsetting ......................................... 56
4.4 Quality assessment ........................................ 57
4.4.1 Diagnostic plots ................................... 57
4.4.2 Spatial plots of spot statistics - image ........... 59
4.4.3 Boxplots of spot statistics - boxplot .............. 60
4.4.4 Scatter-plots of spot statistics - plot ............ 61
4.5 Normalization ............................................. 62
4.5.1 Two-channel normalization .......................... 63
4.5.2 Separate-channel normalization ..................... 64
4.6 Case study ........................................... 67
5 Cell-Based Assays ......................................... 71
W. Huber and F. Hahne
5.1 Scope ..................................................... 71
5.2 Experimental technologies ................................. 71
5.2.1 Expression assays .................................. 72
5.2.2 Loss of function assays ............................ 72
5.2.3 Monitoring the response ............................ 72
5.3 Reading data .............................................. 73
5.3.1 Plate reader data .................................. 74
5.3.2 Further directions in normalization ................ 76
5.3.3 FCS format ......................................... 77
5.4 Quality assessment and visualization ...................... 79
5.4.1 Visualization at the level of individual cells ..... 79
5.4.2 Visualization at the level of microtiter plates .... 82
5.4.3 Brushing with Rggobi ............................... 83
5.5 Detection of effectors .................................... 85
5.5.1 Discrete Response .................................. 85
5.5.2 Continuous response ................................ 88
5.5.3 Outlook ............................................ 90
6 SELDI-TOF Mass Spectrometry Protein Data .................. 91
X. Li, R. Gentleman, X. Lu, Q. Shi, J.D. Iglehart,
L. Harris, and A. Miron
6.1 Introduction .............................................. 91
6.2 Baseline subtraction ...................................... 93
6.3 Peak detection ............................................ 95
6.4 Processing a set of calibration spectra ................... 96
6.4.1 Apply baseline subtraction to a set of spectra ..... 98
6.4.2 Normalize spectra .................................. 99
6.4.3 Cutoff selection .................................. 100
6.4.4 Identify peaks .................................... 101
6.4.5 Quality assessment ................................ 101
6.4.6 Get proto-biomarkers .............................. 102
6.5 An example ............................................... 105
6.6 Conclusion ............................................... 108
II Meta-data: biological annotation and visualiza-
tion ..................................................... 111
7 Meta-data Resources and Tools in Bioconductor ............ 113
R. Gentleman, V.J. Carey, and J. Zhang
7.1 Introduction ............................................. 113
7.2 External annotation resources ............................ 115
7.3 Bioconductor annotation concepts: curated persistent
packages and Web services ................................ 116
7.3.1 Annotating a platform: HG-U95Av2 .................. 117
7.3.2 An Example ........................................ 118
7.3.3 Annotating a genome ............................... 119
7.4 The annotate package ..................................... 119
7.5 Software tools for working with Gene Ontology (GO) ....... 120
7.5.1 Basics of working with the GO package ............. 121
7.5.2 Navigating the hierarchy .......................... 122
7.5.3 Searching for terms ............................... 122
7.5.4 Annotation of GO terms to LocusLink sequences:
evidence codes .................................... 123
7.5.5 The GO graph associated with a term ............... 125
7.6 Pathway annotation packages: KEGG and cMAP ............... 125
7.6.1 KEGG .............................................. 126
7.6.2 cMAP .............................................. 127
7.6.3 A Case Study ...................................... 129
7.7 Cross-organism annotation: the homology packages ......... 130
7.8 Annotation from other sources ............................ 132
7.9 Discussion ............................................... 133
8 Querying On-line Resources ............................... 135
V.J. Carey, D. Temple Lang, J. Gentry, J. Zhang, and
R. Gentleman
8.1 The Tools ................................................ 135
8.1.1 Entrez ............................................ 137
8.1.2 Entrez examples ................................... 137
8.2 PubMed ................................................... 138
8.2.1 Accessing PubMed information ...................... 139
8.2.2 Generating HTML output for your abstracts ......... 141
8.3 KEGG via SOAP ............................................ 142
8.4 Getting gene sequence information ........................ 144
8.5 Conclusion ............................................... 145
9 Interactive Outputs ...................................... 147
C.A. Smith, W. Huber, and R. Gentleman
9.1 Introduction ............................................. 147
9.2 A simple approach ........................................ 148
9.3 Using the annaffy package ................................ 149
9.4 Linking to On-line Databases ............................. 152
9.5 Building HTML pages ...................................... 153
9.5.1 Limiting the results .............................. 153
9.5.2 Annotating the probes ............................. 154
9.5.3 Adding other data ................................. 155
9.6 Graphical displays with drill-down functionality ......... 156
9.6.1 HTML image maps ................................... 157
9.6.2 Scalable Vector Graphics (SVG) .................... 158
9.7 Searching Meta-data ...................................... 159
9.7.1 Text searching .................................... 159
9.8 Concluding Remarks ....................................... 160
10 Visualizing Data ......................................... 161
W. Huber, X. Li, and R. Gentleman
10.1 Introduction ............................................. 161
10.2 Practicalities ........................................... 162
10.3 High-volume scatterplots ................................. 163
10.3.1 A note on performance ............................. 164
10.4 Heatmaps ................................................. 166
10.4.1 Heatmaps of residuals ............................. 168
10.5 Visualizing distances .................................... 170
10.5.1 Multidimensional scaling .......................... 173
10.6 Plotting along genomic coordinates ....................... 174
10.6.1 Cumulative Expression ............................. 178
10.7 Conclusion ............................................... 179
III Statistical analysis for genomic experiments ............. 181
11 Analysis Overview ........................................ 183
V.J. Carey and R. Gentleman
11.1 Introduction and road map ................................ 183
11.1.1 Distance concepts ................................. 184
11.1.2 Differential expression ........................... 184
11.1.3 Cluster analysis .................................. 184
11.1.4 Machine learning .................................. 184
11.1.5 Multiple comparisons .............................. 185
11.1.6 Workflow support .................................. 185
11.2 Absolute and relative expression measures ................ 185
12 Distance Measures in DNA Microarray Data Analysis ........ 189
R. Gentleman, B. Ding, S. Dudoit, and J. Ibrahim
12.1 Introduction ............................................. 189
12.2 Distances ................................................ 191
12.2.1 Definitions ....................................... 191
12.2.2 Distances between points .......................... 192
12.2.3 Distances between distributions ................... 195
12.2.4 Experiment-specific distances between genes ....... 198
12.3 Microarray data .......................................... 199
12.3.1 Distances and standardization ..................... 199
12.4 Examples ................................................. 201
12.4.1 A co-citation example ............................. 203
12.4.2 Adjacency ......................................... 207
12.5 Discussion ............................................... 208
13 Cluster Analysis of Genomic Data ........................ 209
K.S. Pollard and M.J. van der Laan
13.1 Introduction ............................................. 209
13.2 Methods .................................................. 210
13.2.1 Overview of clustering algorithms ................. 210
13.2.2 Ingredients of a clustering algorithm ............. 211
13.2.3 Building sequences of clustering results .......... 211
13.2.4 Visualizing clustering results .................... 214
13.2.5 Statistical issues in clus'tering ................. 215
13.2.6 Bootstrapping a cluster analysis .................. 216
13.2.7 Number of clusters ................................ 217
13.3 Application: renal cell cancer ........................... 222
13.3.1 Gene selection .................................... 222
13.3.2 HOPACH clustering of genes ........................ 223
13.3.3 Comparison with PAM ............................... 224
13.3.4 Bootstrap resampling .............................. 224
13.3.5 HOPACH clustering of arrays ....................... 224
13.3.6 Output files ...................................... 226
13.4 Conclusion ............................................... 228
14 Analysis of Differential Gene Expression Studies ......... 229
D. Scholtens and A. von Heydebreck
14.1 Introduction ............................................. 229
14.2 Differential expression analysis ......................... 230
14.2.1 Example: ALL data ................................. 232
14.2.2 Example: Kidney cancer data ....................... 236
14.3 Multifactor experiments .................................. 239
14.3.1 Example: Estrogen data ............................ 241
14.4 Conclusion .......................................... 248
15 Multiple Testing Procedures: the multtest Package and
Applications to Genomics ................................. 249
K.S. Pollard, S. Dudoit, and M.J. van der Laan
15.1 Introduction ............................................. 249
15.2 Multiple hypothesis testing methodology .................. 250
15.2.1 Multiple hypothesis testing framework ............. 250
15.2.2 Test statistics null distribution ................. 255
15.2.3 Single-step procedures for controlling general
Type I error rates 6{FVri) ........................ 256
15.2.4 Step-down procedures for controlling the
family-wise error rate ............................ 257
15.2.5 Augmentation multiple testing procedures for
controlling tail probability error rates .......... 258
15.3 Software implementation: R multtest package .............. 259
15.3.1 Resampling-based multiple testing procedures:
MTP function ...................................... 260
15.3.2 Numerical and graphical summaries ................. 262
15.4 Applications: ALL microarray data set .................... 262
15.4.1 ALL data package and initial gene filtering ....... 262
15.4.2 Association of expression measures and tumor
cellular subtype: Two-sample ^-statistics ......... 263
15.4.3 Augmentation procedures ........................... 265
15.4.4 Association of expression measures and tumor
molecular subtype: Multi-sample F-statistics ...... 266
15.4.5 Association of expression measures and time to
relapse: Cox t-statistics ......................... 268
15.5 Discussion .............................................. 270
16 Machine Learning Concepts and Tools for Statistical
Genomics ................................................. 273
V.J. Carey
16.1 Introduction 273
16.2 Illustration: Two continuous features; decision regions .. 274
16.3 Methodological issues .................................... 276
16.3.1 Families of learning methods ...................... 276
16.3.2 Model assessment .................................. 281
16.3.3 Metatheorems on learner and feature selection ..... 283
16.3.4 Computing interfaces .............................. 284
16.4 Applications ............................................. 285
16.4.1 Exploring and comparing classifiers with the ALL
data .............................................. 285
16.4.2 Neural net initialization, convergence, and
tuning ............................................ 287
16.4.3 Other methods ..................................... 287
16.4.4 Structured cross-validation support ............... 288
16.4.5 Assessing variable importance ..................... 289
16.4.6 Expression density diagnostics .................... 289
16.5 Conclusions .............................................. 291
17 Ensemble Methods of Computational Inference .............. 293
T. Hothorn, and Bühlmann
17.1 Introduction ............................................. 293
17.2 Bagging and random forests ............................... 295
17.3 Boosting ................................................. 296
17.4 Multiclass problems ...................................... 298
17.5 Evaluation ............................................... 298
17.6 Applications: tumor prediction ........................... 300
17.6.1 Acute lymphoblastic leukemia ...................... 300
17.6.2 Renal cell cancer ................................. 303
17.7 Applications: Survival analysis .......................... 307
17.8 Conclusion ............................................... 310
18 Browser-based Affymetrix Analysis and Annotation ......... 313
С.A. Smith
18.1 Introduction ............................................. 313
18.1,1 Key user interface features ....................... 314
18.2 Deploying webbioc ........................................ 315
18.2.1 System requirements ............................... 315
18.2.2 Installation ...................................... 315
18.2.3 Configuration ..................................... 316
18.3 Using webbioc ............................................ 317
18.3.1 Data Preprocessing ................................ 317
18.3.2 Differential expression multiple testing .......... 318
18.3.3 Linked annotation meta-data ....................... 320
18.3.4 Retrieving results ................................ 321
18.4 Extending webbioc ........................................ 322
18.4.1 Architectural overview ............................ 322
18.4.2 Creating a new module ............................. 324
18.5 Conclusion ............................................... 326
IV Graphs and networks ...................................... 327
19 Introduction and Motivating Examples ..................... 329
R. Gentleman, W. Huber, and V.J. Carey
19.1 Introduction ............................................. 329
19.2 Practicalities ........................................... 330
19.2.1 Representation .................................... 330
19.2.2 Algorithms ........................................ 330
19.2.3 Data Analysis ..................................... 331
19.3 Motivating examples ...................................... 331
19.3.1 Biomolecular Pathways ............................. 331
19.3.2 Gene ontology: A graph of concept-terms ........... 333
19.3.3 Graphs induced by literature references and
citations ......................................... 334
19.4 Discussion ............................................... 336
20 Graphs ................................................... 337
W. Huber, R. Gentleman, and V.J. Carey
20.1 Overview ................................................. 337
20.2 Definitions .............................................. 338
20.2.1 Special types of graphs ........................... 341
20.2.2 Random graphs ..................................... 343
20.2.3 Node and edge labeling ............................ 344
20.2.4 Searching and related algorithms .................. 344
20.3 Cohesive subgroups ....................................... 344
20.4 Distances ................................................ 346
21 Bioconductor Software for Graphs ......................... 347
V.J. Carey, R. Gentleman, W. Huber, J. Centry
21.1 Introduction ............................................. 347
21.2 The graph package ........................................ 348
21.2.1 Getting started ................................... 349
21.2.2 Random graphs ..................................... 352
21.3 The RBGL package ......................................... 352
21.3.1 Connected graphs .................................. 355
21.3.2 Paths and related concepts ........................ 357
21.3.3 RBGL summary ...................................... 360
21.4 Drawing graphs ........................................... 360
21.4.1 Global attributes ................................. 363
21.4.2 Node and edge attributes .......................... 363
21.4.3 The function agopen and the Ragraph class ......... 365
21.4.4 User-defined drawing functions .................... 366
21.4.5 Image maps on graphs .............................. 368
22 Case Studies Using Graphs on Biological Data ............. 369
R. Gentleman, D. Scholtens, B. Ding, V.J. Carey, and
W. Huber
22.1 Introduction ............................................. 369
22.2 Comparing the transcriptome and the interactome .......... 370
22.2.1 Testing associations .............................. 371
22.2.2 Data analysis ..................................... 373
22.3 Using GO ................................................. 374
22.3.1 Finding interesting GO terms ...................... 375
22.4 Literature co-citation ................................... 378
22.4.1 Statistical development ........................... 380
22.4.2 Comparisons of interest ........................... 382
22.4.3 Examples .......................................... 382
22.5 Pathways ................................................. 387
22.5.1 The graph structure of pathways ................... 388
22.5.2 Relating expression data to pathways .............. 390
22.6 Concluding remarks ....................................... 393
V Case studies ............................................. 395
23 Limma: Linear Models for Microarray Data ................. 397
G.K. Smyth
23.1 Introduction ............................................. 397
23.2 Data representations ..................................... 398
23.3 Linear models ............................................ 399
23.4 Simple comparisons ....................................... 400
23.5 Technical Replication .................................... 403
23.6 Within-array replicate spots ............................. 406
23.7 Two groups ............................................... 407
23.8 Several groups ........................................... 409
23.9 Direct two-color designs ................................. 411
23.10 Factorial designs ....................................... 412
23.11 Time course experiments ................................. 414
23.12 Statistics for differential expression .................. 415
23.13 Fitted model objects .................................... 417
23.14 Preprocessing considerations ............................ 418
23.15 Conclusion .............................................. 420
24 Classification with Gene Expression Data ................. 421
M. Dettling
24.1 Introduction ............................................. 421
24.2 Reading and customizing the data ......................... 422
24.3 Training and validating classifiers ...................... 423
24.4 Multiple random divisions ................................ 426
24.5 Classification of test data .............................. 428
24.6 Conclusion ............................................... 429
25 From CEL Files to Annotated Lists of Interesting Genes ... 431
R.A. Irizarry
25.1 Introduction ............................................. 431
25.2 Reading CEL files ........................................ 432
25.3 Preprocessing ............................................ 432
25.4 Ranking and filtering genes .............................. 433
25.4.1 Summary statistics and tests for ranking .......... 434
25.4.2 Selecting cutoffs ................................. 437
25.4.3 Comparison ........................................ 437
25.5 Annotation ............................................... 438
25.5.1 PubMed abstracts .................................. 439
25.5.2 Generating reports ................................ 441
25.6 Conclusion ............................................... 442
A Details on selected resources ............................ 443
A.l Data sets ................................................ 443
A.l.l ALL ............................................... 443
A.1.2 Renal cell cancer ................................. 443
A.1.3 Estrogen receptor stimulation ..................... 443
A.2 URLs for projects mentioned .............................. 444
References .................................................... 445
Index ......................................................... 465
|
