| Applied Statistics | p. 1 |
| Descriptive statistics | p. 1 |
| Frequency distribution | p. 2 |
| Central tendency and variability | p. 2 |
| Correlation | p. 4 |
| Inferential statistics | p. 6 |
| Probability distribution | p. 6 |
| Central limit theorem and normal distribution | p. 7 |
| Statistical hypothesis testing | p. 7 |
| Two-sample t-test | p. 9 |
| Nonparametric test | p. 9 |
| One-factor ANOVA and F-test | p. 10 |
| Simple linear regression | p. 11 |
| Chi-square test of contingency | p. 13 |
| Statistical power analysis | p. 14 |
| DNA Methylation Microarrays and Quality Control | p. 17 |
| DNA methylation microarrays | p. 18 |
| Workflow of methylome experiment | p. 21 |
| Restriction enzyme-based enrichment | p. 21 |
| Immunoprecipitation-based enrichment | p. 21 |
| Image analysis | p. 23 |
| Visualization of raw data | p. 26 |
| Reproducibility | p. 26 |
| Positive and negative controls by exogenous sequences | p. 32 |
| Intensity fold-change and p-value | p. 32 |
| DNA unmethylation profiling | p. 33 |
| Correlation of intensities between tiling arrays | p. 33 |
| Experimental Design | p. 35 |
| Goals of experiment | p. 36 |
| Class comparison and class prediction | p. 36 |
| Class discovery | p. 36 |
| Reference design | p. 37 |
| Dye swaps | p. 39 |
| Balanced block design | p. 39 |
| Loop design | p. 41 |
| Factorial design | p. 42 |
| Time course experimental design | p. 47 |
| How many samples/arrays are needed? | p. 49 |
| Biological versus technical replicates | p. 49 |
| Statistical power analysis | p. 49 |
| Pooling biological samples | p. 55 |
| Appendix | p. 56 |
| Data Normalization | p. 59 |
| Measure of methylation | p. 59 |
| The need for normalization | p. 61 |
| Strategy for normalization | p. 62 |
| Two-color CpG island microarray normalization | p. 63 |
| Global dependence of log methylation ratios | p. 64 |
| Dependence of log ratios on intensity | p. 65 |
| Dependence of log ratios on print-tips | p. 67 |
| Normalized Cy3- and Cy5-intensities | p. 70 |
| Between-array normalization | p. 71 |
| Oligonucleotide arrays normalization | p. 72 |
| Background correction: PM - MM? | p. 72 |
| Quantile normalization | p. 73 |
| Probeset summarization | p. 75 |
| Normalization using control sequences | p. 76 |
| Appendix | p. 79 |
| Significant Differential Methylation | p. 81 |
| Fold change | p. 81 |
| Linear model for log-ratios or log-intensities | p. 84 |
| Microarrays reference design or oligonucleotide chips | p. 84 |
| Sequence-specific dye effect in two-color microarrays | p. 87 |
| t-test for contrasts | p. 88 |
| F-test for joint contrasts | p. 89 |
| P-value adjustment for multiple testing | p. 92 |
| Bonferroni correction | p. 92 |
| False discovery rate | p. 92 |
| Modified t- and F-test | p. 94 |
| Significant variation within and between groups | p. 95 |
| Within-group variation | p. 95 |
| Between-group variation | p. 96 |
| Significant correlation with a co-variate | p. 97 |
| Permutation test for bisulfite sequence data | p. 100 |
| Euclidean distance | p. 101 |
| Entropy | p. 102 |
| Missing data values | p. 103 |
| Appendix | p. 104 |
| Factorial design | p. 104 |
| Time-course experiments | p. 105 |
| Balanced block design | p. 106 |
| Loop design | p. 107 |
| High-Density Genomic Tiling Arrays | p. 109 |
| Normalization | p. 110 |
| Intra- and interarray normalization | p. 110 |
| Sequence-based probe effects | p. 110 |
| Wilcoxon test in a sliding window | p. 112 |
| Probe score or scan statistic | p. 116 |
| False positive rate | p. 116 |
| Boundaries of methylation regions | p. 118 |
| Multiscale analysis by wavelets | p. 119 |
| Unsupervised segmentation by hidden Markov model | p. 121 |
| Principal component analysis and biplot | p. 125 |
| Cluster Analysis | p. 129 |
| Measure of dissimilarity | p. 129 |
| Dimensionality reduction | p. 130 |
| Hierarchical clustering | p. 133 |
| Bottom-up approach | p. 133 |
| Top-down approach | p. 136 |
| K-means clustering | p. 139 |
| Model-based clustering | p. 141 |
| Quality of clustering | p. 142 |
| Statistically significance of clusters | p. 144 |
| Reproducibility of clusters | p. 146 |
| Repeated measurements | p. 146 |
| Statistical Classification | p. 149 |
| Feature selection | p. 149 |
| Discriminant function | p. 152 |
| Linear discriminant analysis | p. 153 |
| Diagonal linear discriminant analysis | p. 154 |
| K-nearest neighbor | p. 154 |
| Performance assessment | p. 155 |
| Leave-one-out cross validation | p. 156 |
| Receiver operating characteristic analysis | p. 159 |
| Interdependency Network of DNA Methylation | p. 163 |
| Graphs and networks | p. 164 |
| Partial correlation | p. 164 |
| Dependence networks from DNA methylation microarrays | p. 165 |
| Network analysis | p. 168 |
| Distribution of connectivities | p. 169 |
| Active epigenetically regulated loci | p. 169 |
| Correlation of connectivities | p. 170 |
| Modularity | p. 171 |
| Time Series Experiment | p. 179 |
| Regulatory networks from microarray data | p. 181 |
| Dynamic model of regulation | p. 182 |
| A penalized likelihood score for parsimonious model | p. 182 |
| Optimization by genetic algorithms | p. 184 |
| Online Annotations | p. 187 |
| Gene centric resources | p. 187 |
| GenBank: A nucleotide sequence database | p. 187 |
| UniGene: An organized view of transcriptomes | p. 188 |
| RefSeq: Reviews of sequences and annotations | p. 188 |
| PubMed: A bibliographic database of biomedical journals | p. 189 |
| dbSNP: Database for nucleotide sequence variation | p. 190 |
| OMIM: A directory of human genes and genetic disorders | p. 190 |
| Entrez Gene: A Web portal of genes | p. 190 |
| PubMeth: A cancer methylation database | p. 192 |
| Gene Ontology | p. 192 |
| Kyoto Encyclopedia of Genes and Genomes | p. 195 |
| UniProt/Swiss-Prot protein knowledgebase | p. 196 |
| The International HapMap Project | p. 198 |
| UCSC human genome browser | p. 198 |
| Public Microarray Data Repositories | p. 205 |
| Epigenetics Society | p. 205 |
| Microarray Gene Expression Data Society | p. 206 |
| Minimum Information about a Microarray Experiment | p. 206 |
| Public repositories for high-throughput arrays | p. 208 |
| Gene Expression Omnibus at NCBI | p. 208 |
| ArrayExpress at EBI | p. 208 |
| Center for Information Biology Gene Expression data-base at DDBJ | p. 210 |
| Open Source Software for Microarray Data Analysis | p. 211 |
| R: A language and environment for statistical computing and graphics | p. 212 |
| Bioconductor | p. 212 |
| Marray package | p. 215 |
| Affy package | p. 215 |
| Limma package | p. 215 |
| Stats package | p. 215 |
| TilingArray package | p. 217 |
| Ringo package | p. 217 |
| Cluster package | p. 217 |
| Class package | p. 217 |
| GeneNet package | p. 217 |
| Inetwork package | p. 217 |
| GOstats package | p. 218 |
| Annotate package | p. 218 |
| References | p. 219 |
| Index | p. 225 |
| Table of Contents provided by Ingram. All Rights Reserved. |
