This book presents an overview of recent developments in biostatistics and bioinformatics. Written by active researchers in these emerging areas, it is intended to give graduate students and new researchers an idea of where the frontiers of biostatistics and bioinformatics are as well as a forum to learn common techniques in use, so that they can advance the fields via developing new techniques and new results. Extensive references are provided so that researchers can follow the threads to learn more comprehensively what the literature is and to conduct their own research. In particulars, the book covers three important and rapidly advancing topics in biostatistics: analysis of survival and longitudinal data, statistical methods for epidemiology, and bioinformatics.
Author(s): Jianqing Fan
Edition: 1
Publisher: World Scientific Publishing Company
Year: 2009
Language: English
Pages: 295
Tags: Медицинские дисциплины;Социальная медицина и медико-биологическая статистика;
Contents......Page 10
Preface......Page 6
Part I Analysis of Survival and Longitudinal Data......Page 14
1 Introduction.......Page 16
2 Cox's type of models......Page 17
2.1 Cox's models with unknown nonlinear risk functions......Page 22
2.2 Partly linear Cox's models......Page 23
2.3 Partly linear additive Cox's models......Page 24
3 Multivariate Cox's type of models.......Page 27
3.1 Marginal modeling using Cox's models with linear risks......Page 28
3.2 Marginal modeling using Cox's models with nonlinear risks......Page 31
3.3 Marginal modeling using partly linear Cox's models......Page 33
3.4 Marginal modeling using partly linear Cox's models with varying coefficients......Page 35
4 Model selection on Cox's models......Page 37
5 Validating Cox's type of models.......Page 40
6 Transformation models.......Page 41
References.......Page 43
1 Introduction.......Page 48
2 Inference procedure and asymptotic properties.......Page 50
3 Assessing additive and accelerated covariates......Page 53
4 Simulation studies.......Page 54
5 Application......Page 55
6 Remarks......Page 56
Appendix......Page 57
References.......Page 61
1 Introduction......Page 62
2 The quadratic inference function approach.......Page 64
2.1 Generalized estimating equations......Page 65
2.2 Quadratic inference functions......Page 66
3.1 Time-varying coefficient models......Page 69
3.2 Variable selection for longitudinal data......Page 71
4.1 Missing data......Page 73
4.2 Outliers and contamination......Page 74
4.3 A real data example......Page 75
5 Further research and concluding remarks.......Page 78
References.......Page 81
1 Introduction......Page 86
2 Basic concepts of spatial process.......Page 89
2.1 Spatial regression models for normal data......Page 91
2.2 Spatial prediction (Kriging)......Page 93
3 Spatial models for non-normal/discrete data......Page 95
3.1 Spatial generalized linear mixed models (SGLMMs)......Page 96
3.2 Computing MLEs for SGLMMs......Page 99
4 Spatial models for censored outcome data......Page 101
4.1 A class of semiparametric estimation equations......Page 103
4.2 Asymptotic Properties and Variance Estimation......Page 106
4.3 A data example: east boston asthma study......Page 107
References.......Page 109
Part II Statistical Methods for Epidemiology......Page 114
1 Introduction......Page 116
2.1 Traditional design......Page 117
2.2 Marker by treatment interaction design......Page 118
2.4 Modified marker-based strategy design......Page 119
2.5 Targeted design......Page 120
3 Test of hypotheses and sample size calculation.......Page 121
3.2 Test for non-inferiority or superiority......Page 122
3.3 Test for equivalence......Page 123
4 Sample size calculation......Page 124
4.1 Traditional design......Page 125
4.2 Marker by treatment interaction design......Page 126
4.3 Marker-based strategy design......Page 127
4.5 Targeted design......Page 128
5.1 Marker by treatment interaction design......Page 129
5.4 Targeted design......Page 130
6 Conclusions......Page 131
Acknowledgements......Page 134
A.I Traditional design......Page 135
A.2 Marker by treatment interaction design......Page 136
A.3 Marker-based strategy design......Page 137
A.4 Modified marker-based strategy design......Page 138
References......Page 139
1 Introduction......Page 140
2 Two-phase case-control or cross-sectional studies.......Page 143
2.1 Estimating-equation approaches for analyzing two-pha se case-control studies......Page 145
2.2 Nonparametric maximum likelihood analysis of two-ph a se case-control studies......Page 147
3 Two-phase designs in cohort studies......Page 149
3.1.1 Weighted likelihood/estimating equation approaches......Page 151
3.1.2 Pseudo-likelihood estimators......Page 154
3.1.3 Nonparametric maximum likelihood estimation......Page 155
3.1.4 Selection of a method for analysis......Page 156
3.2 Nested case-control and counter-matching design......Page 158
3.2.1 Methods for analyzing the nested case-control data......Page 159
3.2.2 Methods for analyzing counter-matched data......Page 160
3.2.3 Unmatched case-control studies......Page 161
4 Conclusions......Page 162
References......Page 164
Part III Bioinformatics......Page 170
1 Introduction......Page 172
2 Data sources useful for protein interaction predictions......Page 174
3.1 Maximum likelihood-based methods (MLE)......Page 176
3.2 Bayesian methods (BAY)......Page 178
3.3 Domain pair exclusion analysis (DPEA)......Page 179
3.4 Parsimony explanation method (PE)......Page 180
4.1 Integrating different types of genomic information......Page 182
4.3 Prediction performance comparison......Page 184
5 Complex detection methods......Page 185
5.2 Graph clustering methods......Page 186
5.3 Performance comparison......Page 187
References......Page 188
1 Introduction......Page 192
2 A Bayesian approach to motif discovery.......Page 194
2.1 Markov chain Monte Carlo computation......Page 195
2.2 Some extensions of the product-multinomial model......Page 196
3 Discovery of regulatory modules.......Page 197
3.1.1 Evolutionary Monte Carlo for module selection......Page 199
3.1.2 Sampling motif sites A through recursive DA......Page 200
3.2 A case-study......Page 201
4 Motif discovery in multiple species.......Page 202
4.1 The coupled hidden Markov model......Page 203
4.2 Gibbs sampling and Bayesian inference......Page 205
4.3 Simulation studies......Page 206
5 Motif learning on ChIP-chip data......Page 208
5.1 Feature extraction......Page 209
5.2 Bayesian additive regression trees......Page 210
5.3 Application to human ChIP-chip data......Page 211
6 Using nucleosome positioning information in motif discovery........Page 214
6.2 Model fitting and parameter estimation......Page 215
7 Conclusion......Page 217
References.......Page 218
1.1 Cancer genomic alterations......Page 222
1.2 Identifying cancer genomic alterations using oligonucleotide SNP microarrays......Page 223
2.2 Tumor-only LOH inference......Page 225
3.1 Obtaining raw copy numbers from SNP array data......Page 229
3.2 Inferring integer copy numbers......Page 230
3.3 Copy number analysis results of the 10K dataset......Page 233
3.4 Allele-specific copy numbers and major copy proportion......Page 234
3.5 Copy number variations in normal samples and other diseases......Page 235
4.1 Finding significantly altered chromosome regions across multiple samples......Page 237
4.2 Hierarchical clustering analysis......Page 240
4.3 Integrating SNP array data with gene expression data......Page 241
5.1 The dChip software for analyzing SNP array data......Page 242
5.2 Other software packages......Page 243
References.......Page 244
1 Background molecular biology.......Page 252
2.1 Chromatin immunoprecipitation......Page 254
2.4 Commercial tiling microarrays......Page 256
3 Data description and analysis.......Page 258
3.1 Low-level analysis......Page 259
3.2 High-level analysis......Page 260
4 Follow-up analysis.......Page 262
4.2 Scanning sequences for known motifs......Page 263
4.3.1 Regular expression enumeration......Page 265
4.3.2 Position weight matrix update......Page 266
References......Page 267
Subject Index.......Page 272
Author Index......Page 274
