DHAMMIKA AMARATUNGA, PhD, is Senior Director and Janssen Fellow in the Nonclinical Statistics and Computing Department at Janssen R&D, a Johnson & Johnson pharmaceutical company. His research interests include analysis of large multivariate data sets generated by functional genomics research, robust and resistant statistical methods, linear and nonlinear modeling, and biostatistics.JAVIER CABRERA, PhD, is Full Professor in the Department of Statistics at Rutgers University. He has published over 100 articles in his areas of research interest, which include DNA microarray, data mining of biopharmaceutical databases, computer vision, statistical computing and graphics, robustness, and biostatistics. He has also lectured at Cold Spring Harbor Laboratory, The Hong Kong University of Science and Technology, and National University of Singapore.ZIV SHKEDY, PhD, is Associate Professor and Statistical Consultant in the Interuniversity Institute for Biostatistics and Statistical Bioinformatics, Center for Statistics at Hasselt University, Belgium. He has published numerous journal articles on the topics of clinical and non-clinical trials, modeling infectious disease data, dose-response analysis, Bayesian modeling, bioinformatics, and analysis of gene expression data.

• Preface xvAcknowledgments xvii1 A brief introduction 11.1 A note on exploratory data analysis 31.2 Computing considerations and software 41.3 A brief outline of the book 51.4 Datasets and case studies 72 Genomics basics 112.1 Genes 112.2 DNA 122.3 Gene expression 132.4 Hybridization assays and other laboratory techniques 152.5 The human genome 162.6 Genome variations and their consequences 182.7 Genomics 192.8 The role of genomics in pharmaceutical and research and clinical practice 202.9 Proteins 232.10 Bioinformatics 233 Microarrays 273.1 Types of microarray experiments 283.2 A very simple hypothetical microarray experiment 323.3 A typical microarray experiment 343.4 Multichannel cDNA microarrays 383.5 Oligonucleotide microarrays 383.6 Bead based arrays 403.7 Confirmation of microarray results 404 Processing the scanned image 434.1 Converting the scanned image to the spotted image 444.2 Quality assessment 474.3 Adjusting for background 534.4 Expression level calculation for twochannel cDNA microarrays 564.5 Expression level calculation for oligonucleotide microarrays 585 Preprocessing microarray data 655.1 Logarithmic transformation 665.2 Variance stabilizing transformations 665.3 Sources of bias 685.4 Normalization 695.5 Intensity dependent normalization 705.6 Judging the success of a normalization 815.7 Outlier identification 835.8 Nonresistant rules for outlier identification 835.9 Resistant rules for outlier identification 835.10 Assessing replicate array quality 846 Summarization 956.1 Replication 956.2 Technical replicates 966.3 Biological replicates 1006.4 Biological replicates 1006.5 Multiple oligonucleotide arrays 1026.6 Estimating fold change in twochannel experiments 1046.7 Bayes estimation of fold change 1056.8 Estimating fold change Affymetrix data 1066.9 RMA Summarization of multiple oligonucleotide arrays revisited 1076.10 FARMS summarization. 1087 Two group comparative experiments 1197.1 Basics of statistical hypothesis testing 1207.2 Fold changes 1237.3 The two sample t test 1237.4 Diagnostic checks 1277.5 Robust t tests 1297.6 The Mann Whitney Wilcox on rank sum test 1307.7 Multiplicity 1327.8 The false discovery rate 1357.9 Resampling based Multiple Testing Procedures 1387.10 Small variance adjusted t tests and SAM 1407.11 Conditional t 1467.12 Borrowing strength across genes 1497.13 Twochannel experiments 1517.14 Filtering 1538 Model based inference and experimental design considerations 1778.1 The F test 1788.2 The basic linear model 1798.3 Fitting the model in two stages 1818.4 Multichannel experiments 1828.5 Experimental design considerations 1838.6 Miscellaneous issues 1878.7 Model based analysis of Affymetrix arrays 1889 Analysis of gene sets 2119.1 Methods for identifying enriched gene sets 2139.2 ORA and Fisher’s exact test 2179.3 Interpretation of results 2179.4 Example 21710 Pattern discovery 22110.1 Initial considerations 22210.2 Cluster analysis 22310.3 Seeking patterns visually 24110.4 Biclustering 25411 Class prediction 26311.1 Initial considerations 26411.2 Linear Discriminant Analysis 26911.3 Extensions of Fisher’s LDA 27511.4 Penalized methods 27811.5 Nearest neighbors 27911.6 Recursive partitioning 28011.7 Ensemble methods 28511.8 Enriched ensemble classifiers 28811.9 Neural networks 28811.10 Support Vector Machines 28911.11 Generalized enriched methods 29111.12 Integration of genome information 30112 Protein arrays 30712.1 Introduction 30712.2 Protein array experiments 30812.3 Special issues with protein arrays 31012.4 Analysis 31012.5 Using antibody antigen arrays to measure protein concentrations 311References 317Index 337

“Featuring new information on interpretation of findings, class prediction, ABC clustering, limma for mixed models, biclustering, mass spectrometry, tracking Spearman correlations, and more, this \extremely well written" (Journal of Environmental Quality) book is a choice reference for scientists, teachers, and students interested in DNA data analysis.”  (Zentralblatt MATH, 1 October 2014)“In summary this is an excellent text for both life scientist and computer/mathematicians.  Highly recommended.”  (Scientific Computing, 1 August 2014)

DHAMMIKA AMARATUNGA, PhD, is Senior Director and Janssen Fellow in the Nonclinical Statistics and Computing Department at Janssen R&D, a Johnson & Johnson pharmaceutical company. His research interests include analysis of large multivariate data sets generated by functional genomics research, robust and resistant statistical methods, linear and nonlinear modeling, and biostatistics.JAVIER CABRERA, PhD, is Full Professor in the Department of Statistics at Rutgers University. He has published over 100 articles in his areas of research interest, which include DNA microarray, data mining of biopharmaceutical databases, computer vision, statistical computing and graphics, robustness, and biostatistics. He has also lectured at Cold Spring Harbor Laboratory, The Hong Kong University of Science and Technology, and National University of Singapore.ZIV SHKEDY, PhD, is Associate Professor and Statistical Consultant in the Interuniversity Institute for Biostatistics and Statistical Bioinformatics, Center for Statistics at Hasselt University, Belgium. He has published numerous journal articles on the topics of clinical and non-clinical trials, modeling infectious disease data, dose-response analysis, Bayesian modeling, bioinformatics, and analysis of gene expression data.

• Preface xvAcknowledgments xvii1 A brief introduction 11.1 A note on exploratory data analysis 31.2 Computing considerations and software 41.3 A brief outline of the book 51.4 Datasets and case studies 72 Genomics basics 112.1 Genes 112.2 DNA 122.3 Gene expression 132.4 Hybridization assays and other laboratory techniques 152.5 The human genome 162.6 Genome variations and their consequences 182.7 Genomics 192.8 The role of genomics in pharmaceutical and research and clinical practice 202.9 Proteins 232.10 Bioinformatics 233 Microarrays 273.1 Types of microarray experiments 283.2 A very simple hypothetical microarray experiment 323.3 A typical microarray experiment 343.4 Multichannel cDNA microarrays 383.5 Oligonucleotide microarrays 383.6 Bead based arrays 403.7 Confirmation of microarray results 404 Processing the scanned image 434.1 Converting the scanned image to the spotted image 444.2 Quality assessment 474.3 Adjusting for background 534.4 Expression level calculation for twochannel cDNA microarrays 564.5 Expression level calculation for oligonucleotide microarrays 585 Preprocessing microarray data 655.1 Logarithmic transformation 665.2 Variance stabilizing transformations 665.3 Sources of bias 685.4 Normalization 695.5 Intensity dependent normalization 705.6 Judging the success of a normalization 815.7 Outlier identification 835.8 Nonresistant rules for outlier identification 835.9 Resistant rules for outlier identification 835.10 Assessing replicate array quality 846 Summarization 956.1 Replication 956.2 Technical replicates 966.3 Biological replicates 1006.4 Biological replicates 1006.5 Multiple oligonucleotide arrays 1026.6 Estimating fold change in twochannel experiments 1046.7 Bayes estimation of fold change 1056.8 Estimating fold change Affymetrix data 1066.9 RMA Summarization of multiple oligonucleotide arrays revisited 1076.10 FARMS summarization. 1087 Two group comparative experiments 1197.1 Basics of statistical hypothesis testing 1207.2 Fold changes 1237.3 The two sample t test 1237.4 Diagnostic checks 1277.5 Robust t tests 1297.6 The Mann Whitney Wilcox on rank sum test 1307.7 Multiplicity 1327.8 The false discovery rate 1357.9 Resampling based Multiple Testing Procedures 1387.10 Small variance adjusted t tests and SAM 1407.11 Conditional t 1467.12 Borrowing strength across genes 1497.13 Twochannel experiments 1517.14 Filtering 1538 Model based inference and experimental design considerations 1778.1 The F test 1788.2 The basic linear model 1798.3 Fitting the model in two stages 1818.4 Multichannel experiments 1828.5 Experimental design considerations 1838.6 Miscellaneous issues 1878.7 Model based analysis of Affymetrix arrays 1889 Analysis of gene sets 2119.1 Methods for identifying enriched gene sets 2139.2 ORA and Fisher’s exact test 2179.3 Interpretation of results 2179.4 Example 21710 Pattern discovery 22110.1 Initial considerations 22210.2 Cluster analysis 22310.3 Seeking patterns visually 24110.4 Biclustering 25411 Class prediction 26311.1 Initial considerations 26411.2 Linear Discriminant Analysis 26911.3 Extensions of Fisher’s LDA 27511.4 Penalized methods 27811.5 Nearest neighbors 27911.6 Recursive partitioning 28011.7 Ensemble methods 28511.8 Enriched ensemble classifiers 28811.9 Neural networks 28811.10 Support Vector Machines 28911.11 Generalized enriched methods 29111.12 Integration of genome information 30112 Protein arrays 30712.1 Introduction 30712.2 Protein array experiments 30812.3 Special issues with protein arrays 31012.4 Analysis 31012.5 Using antibody antigen arrays to measure protein concentrations 311References 317Index 337

“Featuring new information on interpretation of findings, class prediction, ABC clustering, limma for mixed models, biclustering, mass spectrometry, tracking Spearman correlations, and more, this \extremely well written" (Journal of Environmental Quality) book is a choice reference for scientists, teachers, and students interested in DNA data analysis.”  (Zentralblatt MATH, 1 October 2014)“In summary this is an excellent text for both life scientist and computer/mathematicians.  Highly recommended.”  (Scientific Computing, 1 August 2014)

DHAMMIKA AMARATUNGA, PhD, is Senior Director and Janssen Fellow in the Nonclinical Statistics and Computing Department at Janssen R&D, a Johnson & Johnson pharmaceutical company. His research interests include analysis of large multivariate data sets generated by functional genomics research, robust and resistant statistical methods, linear and nonlinear modeling, and biostatistics.JAVIER CABRERA, PhD, is Full Professor in the Department of Statistics at Rutgers University. He has published over 100 articles in his areas of research interest, which include DNA microarray, data mining of biopharmaceutical databases, computer vision, statistical computing and graphics, robustness, and biostatistics. He has also lectured at Cold Spring Harbor Laboratory, The Hong Kong University of Science and Technology, and National University of Singapore.ZIV SHKEDY, PhD, is Associate Professor and Statistical Consultant in the Interuniversity Institute for Biostatistics and Statistical Bioinformatics, Center for Statistics at Hasselt University, Belgium. He has published numerous journal articles on the topics of clinical and non-clinical trials, modeling infectious disease data, dose-response analysis, Bayesian modeling, bioinformatics, and analysis of gene expression data.

• Preface xvAcknowledgments xvii1 A brief introduction 11.1 A note on exploratory data analysis 31.2 Computing considerations and software 41.3 A brief outline of the book 51.4 Datasets and case studies 72 Genomics basics 112.1 Genes 112.2 DNA 122.3 Gene expression 132.4 Hybridization assays and other laboratory techniques 152.5 The human genome 162.6 Genome variations and their consequences 182.7 Genomics 192.8 The role of genomics in pharmaceutical and research and clinical practice 202.9 Proteins 232.10 Bioinformatics 233 Microarrays 273.1 Types of microarray experiments 283.2 A very simple hypothetical microarray experiment 323.3 A typical microarray experiment 343.4 Multichannel cDNA microarrays 383.5 Oligonucleotide microarrays 383.6 Bead based arrays 403.7 Confirmation of microarray results 404 Processing the scanned image 434.1 Converting the scanned image to the spotted image 444.2 Quality assessment 474.3 Adjusting for background 534.4 Expression level calculation for twochannel cDNA microarrays 564.5 Expression level calculation for oligonucleotide microarrays 585 Preprocessing microarray data 655.1 Logarithmic transformation 665.2 Variance stabilizing transformations 665.3 Sources of bias 685.4 Normalization 695.5 Intensity dependent normalization 705.6 Judging the success of a normalization 815.7 Outlier identification 835.8 Nonresistant rules for outlier identification 835.9 Resistant rules for outlier identification 835.10 Assessing replicate array quality 846 Summarization 956.1 Replication 956.2 Technical replicates 966.3 Biological replicates 1006.4 Biological replicates 1006.5 Multiple oligonucleotide arrays 1026.6 Estimating fold change in twochannel experiments 1046.7 Bayes estimation of fold change 1056.8 Estimating fold change Affymetrix data 1066.9 RMA Summarization of multiple oligonucleotide arrays revisited 1076.10 FARMS summarization. 1087 Two group comparative experiments 1197.1 Basics of statistical hypothesis testing 1207.2 Fold changes 1237.3 The two sample t test 1237.4 Diagnostic checks 1277.5 Robust t tests 1297.6 The Mann Whitney Wilcox on rank sum test 1307.7 Multiplicity 1327.8 The false discovery rate 1357.9 Resampling based Multiple Testing Procedures 1387.10 Small variance adjusted t tests and SAM 1407.11 Conditional t 1467.12 Borrowing strength across genes 1497.13 Twochannel experiments 1517.14 Filtering 1538 Model based inference and experimental design considerations 1778.1 The F test 1788.2 The basic linear model 1798.3 Fitting the model in two stages 1818.4 Multichannel experiments 1828.5 Experimental design considerations 1838.6 Miscellaneous issues 1878.7 Model based analysis of Affymetrix arrays 1889 Analysis of gene sets 2119.1 Methods for identifying enriched gene sets 2139.2 ORA and Fisher’s exact test 2179.3 Interpretation of results 2179.4 Example 21710 Pattern discovery 22110.1 Initial considerations 22210.2 Cluster analysis 22310.3 Seeking patterns visually 24110.4 Biclustering 25411 Class prediction 26311.1 Initial considerations 26411.2 Linear Discriminant Analysis 26911.3 Extensions of Fisher’s LDA 27511.4 Penalized methods 27811.5 Nearest neighbors 27911.6 Recursive partitioning 28011.7 Ensemble methods 28511.8 Enriched ensemble classifiers 28811.9 Neural networks 28811.10 Support Vector Machines 28911.11 Generalized enriched methods 29111.12 Integration of genome information 30112 Protein arrays 30712.1 Introduction 30712.2 Protein array experiments 30812.3 Special issues with protein arrays 31012.4 Analysis 31012.5 Using antibody antigen arrays to measure protein concentrations 311References 317Index 337

“Featuring new information on interpretation of findings, class prediction, ABC clustering, limma for mixed models, biclustering, mass spectrometry, tracking Spearman correlations, and more, this \extremely well written" (Journal of Environmental Quality) book is a choice reference for scientists, teachers, and students interested in DNA data analysis.”  (Zentralblatt MATH, 1 October 2014)“In summary this is an excellent text for both life scientist and computer/mathematicians.  Highly recommended.”  (Scientific Computing, 1 August 2014)