Genetic Dissection of Complex Traits (eBook)

Cover 1
Contents 6
Contributors 16
Preface to First Edition 22
Preface to Second Edition 26
Acknowledgments 28
Part I: Overview and Fundamentals 29
Chapter 1: An Overview of the Genetic Dissection of Complex Traits 31
I. Introduction 32
II. Challenges Arising From Complex Traits 35
III. Study Design 37
A. Phenotype 37
B. Genotype 38
C. Linkage versus association 38
D. Sampling type 39
E. Sample size, significance level, and power 39
IV. Analytical Methods 40
A. Familial aggregation and genetic effects 41
B. Linkage, association, and admixture mapping 41
C. Dense SNPs and haplotype analysis 44
D. Composite likelihoods 46
E. Multiple testing 46
V. Special Topics 47
A. Pathway-based association studies and Bayesian networks 47
B. Gene expression and systems biology 48
C. Comparative genomics 49
VI. GWA Studies 50
A. Recent GWA studies 50
B. Designing GWA studies 51
C. On transferability of genome-wide tagSNPs 52
D. Follow-up studies 53
VII. Efficient Strategies for Enhancing Gene Discovery 53
A. Lumping and splitting 53
B. Meta-analysis 55
C. Multivariate phenotypes 56
VIII. Discussion 56
Acknowledgments 57
References 58
Chapter 2: Familial Resemblance and Heritability 63
I. Introduction 64
II. Familial Resemblance and Heritability 66
A. Family resemblance 66
B. Heritability 66
C. Risk-based estimates of heritability 67
III. Study Designs and Multifactorial Models 69
A. Nuclear families 69
B. Extended pedigrees 71
C. Twins 71
D. Adoptions 71
E. Modeling extensions 72
F. Factors affecting heritability estimation 73
IV. Discussion 74
Acknowledgments 75
References 75
Chapter 3: Linkage and Association: Basic Concepts 79
I. Introduction 80
II. Historical Perspective 82
III. Fundamentals and Methods 83
A. Linkage analysis 83
B. Association analysis 87
IV. Seven Stages of Relationship With GW Studies 92
V. Contemporary Approaches 94
VI. Challenges and Issues 97
A. Multiple comparisons and type I and type II errors 97
B. Genetic heterogeneity 97
Acknowledgments 98
References 98
Chapter 4: Definition of Phenotype 103
I. Introduction 104
II. Phenotype Choices/Options 106
A. Discrete versus continuous traits 106
B. Study base differences 106
C. Clinical heterogeneity 109
D. Endophenotype 115
E. Composite phenotype 116
F. Narrow versus broad phenotype definition 120
III. Study Design Issues 120
A. Measurement error 120
B. Misclassification 121
C. Loss due to dichotomization/categorization 125
D. Effect of narrowing phenotype 126
E. Multiple comparisons 127
IV. Discussion 127
Acknowledgments 128
References 128
Chapter 5: Genotyping Platforms for Mass-Throughput Genotyping with SNPs, Including Human Genome-Wide Scans 135
I. Introduction: Explosive Need for Mass-Throughput Genotyping 136
II. Microsatellites Versus SNPs 138
A. Microsatellites 138
B. SNPs 139
C. SNPs in whole-genome association studies 140
D. SNPs in whole-genome linkage scans 141
III. Genotyping Platforms and Goals 143
A. Challenges to mass-throughput genotyping 143
B. Genome-Wide platforms 145
C. Targeted genotyping platforms 151
IV. Selection of Markers and SNPs 154
A. Selection of genome-wide platforms 154
B. Selection of targeted platforms 158
V. Novel Markers 158
VI. Future Prospects 161
A. SNP-typing platforms in the clinic 162
B. Integrative array-based genomic analyses 163
VII. Conclusions 163
References 164
Chapter 6: Genotyping Errors and Their Impact on Genetic Analysis 169
I. Introduction 170
II. Misspecification of Genetic Relationships 171
A. Dissociation of marker data from the individual ID 171
B. Inaccurate information about familial relations 171
C. Errors in data entry 172
D. Related families specified as unrelated 172
III. Detection and Correction of Misspecification of Relationships 172
A. GRR 172
B. ASPEX 175
C. Eclipse 175
IV. Genotype Misclassification 175
A. Misclassification rates 176
B. Mendelian inconsistencies 176
C. Unlikely double recombinants 177
D. Allele shifting 178
V. Conclusion 179
References 179
Part II: Linkage and Association Analysis 181
Chapter 7: Model-Based Methods for Linkage Analysis 183
I. Introduction 184
II. The Generalized Single Major Locus Model Description 184
III. The Lod Score 185
A. Definition 185
B. Examples 185
C. The LOD score of 3 criterion 188
D. Genome-wide significance level 189
E. Affected sib pair methods and LOD scores 190
F. Maximum LOD score for ASP 191
G. Maximum LOD score versus the LOD score 192
IV. The Mod Score Method 193
V. The Lod Score and Meta-Analysis 194
A. SML traits 194
B. Complex traits 195
VI. Multilocus Models 196
VII. Strengths and Weaknesses of Model-Based Methods 198
VIII. Discussion 198
References 199
Chapter 8: Contemporary Model-Free Methods for Linkage Analysis 203
I. Introduction 204
II. Identity by Descent and Identity by State 205
A. Multipoint IBD 207
B. IBD estimation and intermarker linkage disequilibrium 208
III. Relative Pair Linkage Methods 208
A. Discordant pairs 210
B. Covariates in affected sib pair analyses 211
C. Power 214
IV. Variance Components Linkage Methods 212
A. Nonnormality of the trait distribution 213
B. Discrete and categorical traits 214
C. Power 214
V. Strengths and Weaknesses of Model-Free Linkage Methods 215
VI. Conclusion 217
Acknowledgments 217
References 217
Chapter 9: DNA Sequence-Based Phenotypic Association Analysis 223
I. Introduction 224
A. DNA sequencing and association studies 226
B. Potential analysis methods 227
II. Multivariate Distance Matrix Regression 230
A. Computing a distance matrix 230
B. MDMR analysis 232
C. Assessing significance of the F statistic 233
D. Graphical display of distance/similarity matrices 234
III. Simulation Studies 234
A. The determination of critical values 235
B. The assessment of the power of MDMR 235
IV. Results 237
A. The determination of critical values 237
B. The assessment of the power of MDMR: Equal effect sizes 239
C. The assessment of the power of MDMR: Diminishing effect sizes 241
D. The assessment of the power of MDMR: The influence of neutral loci 241
V. Discussion 241
Acknowledgments 243
References 243
Chapter 10: Family-Based Methods for Linkage and Association Analysis 247
I. Introduction 248
A. Hypothesis testing in family designs 248
B. The TDT test for trios 250
C. Extensions to the TDT 251
D. Design issues 252
II. Analysis Methods: FBAT and PBAT 256
A. General test statistic 258
B. Coding the genotype 258
C. Coding the trait: Dichotomous outcomes 259
D. The test statistic: Large sample distribution under the null 259
E. The TDT and chi2FBAT 260
F. Computing the distribution with general pedigrees and/or missing founders 261
G. Haplotypes and multiple markers 262
H. Coding the trait for complex phenotypes: Age-to-onset phenotypes, quantitative outcomes, and FBAT-GEE 264
I. A General approach to complex phenotypes: Separating the population and family information in family data 266
J. Testing strategies for large-scale association studies 268
III. Other Approaches to Family-Based Analyses, Including the PDT and the QTDT 268
A. The PDT and APL 272
B. Quantitative traits: The QTDT 273
IV. Software 274
V. Discussion 274
References 276
Chapter 11: Searching for Additional Disease Loci in a Genomic Region 281
I. Introduction 282
A. Background 282
B. Primary disease genes 285
C. Additional (secondary) disease genes in a genetic region 286
II. Primary Disease-Predisposing Genes and Their Genetic Features 289
A. Background 289
B. Linkage and association tests 290
C. Modes of inheritance (recessive versus additive models) 292
D. The patient/control ratio 295
E. Relative predispositional effects 296
III. Detecting Additional Genes in a Genetic Region 299
A. Background 299
B. Linkage analyses of two linked disease susceptibility genes 299
C. Linkage disequilibrium of markers with primary disease genes 300
D. Matched cases and controls 301
E. Homozygous parent linkage and TDT methods 302
F. Conditional haplotype methods 303
G. Conditional genotype methods 307
I. Combining association and IBD values 309
IV. Discussion 311
Acknowledgments 313
References 313
Chapter 12: Methods for Handling Multiple Testing 321
I.Introduction 322
II. Types of Errors in Hypothesis Testing (Type I and Type II) 324
III. Striking a Balance Between False Positives and False Negatives 326
IV. Alternative Adjustment Methods 327
A. The Bonferroni correction 327
B. Permutation testing 328
C. False discovery rate 329
D. Sequential methods 330
E. Other methods 332
V. Conclusion 333
Acknowledgments 333
References 333
Part III: Special Topics 337
Chapter 13: Meta-Analysis Methods 339
I. Introduction 340
II. Meta-Analysis of Population-Based Association Studies 341
A. Background considerations 341
B. Special issues in genetic applications 346
III. Meta-Analysis of Linkage Studies 347
A. Meta-analysis of significance levels 348
B. Parametric meta-analysis for linkage studies 350
C. Meta-analysis of genome scans 350
IV. Special Issues 352
A. Nonreplication of early genetic claims 352
B. Biases in meta-analyses, with emphasis on genetic meta-analysis 354
C. Meta-analyses of individual participant data and consortia of investigators 355
D. Meta-analysis of genome-wide association studies 355
E. Meta-analysis of gene-gene-environment data 356
References 357
Chapter 14: Haplotype-Association Analysis 363
I. Introduction 364
II. Haplotype Inference From Unrelated Individuals 365
A. Statistical methods 366
B. Combinatorial algorithms 370
C. Some extensions and other methods 373
III. Haplotype Inference From Pedigrees 377
A. Brief introduction for haplotype inference using pedigrees 377
B. Rule-based methods for haplotype inference using pedigrees 378
C. Likelihood-based methods for haplotype inference using pedigrees 383
D. The program packages 390
IV. Population-Based Haplotype-Association Methods 390
A. Score statistics for haplotype analysis 390
B. Haplotype-association analysis using regression models 395
V. Family-Based Association Methods Using Haplotypes 404
A. Brief introduction 404
B. Family-based association methods using single markers 405
C. Methods for phase known data 407
D. Methods for phase unknown data 410
E. The program packages 419
VI. Discussion 419
Acknowledgments 421
References 421
Chapter 15: Characterization of LD Structures and the Utility of HapMap in Genetic Association Studies 435
I. Introduction 436
II. Haplotype Similarity in Candidate Genes and LD Mapping 437
A. Founder heterogeneity 438
B. Multiple candidate genes and global test 439
C. Other practical issues 441
III. Global Organization of LD Structures 442
A. HapMap and tagSNPs 443
B. Similarity of HapMaps and transferability of tagSNPs 446
C. Implications to GWA 446
IV. Variation of Local LD Structures 448
A. The marker ambiguity score method 449
B. Implications for GWA studies 451
V. Discussion 453
Acknowledgment 457
References 458
Chapter 16: Associations Among Multiple Markers and Complex Disease: Models, Algorithms, and Applications 465
I. Introduction 466
II. A Model for "Diplotypes" 632
A. Independence and the multinomial model 470
B. Strand-specific probabilities and resampling 470
III. Strand-Specific Inference in the Absence of Randomness in P 474
IV. Haploview and Our View 632
A. Testing: Individual pairs of alleles and omnibus tests 476
B. Clustering 480
V. Approaches to Two-Class Association/Classification 482
VI. Restrictions on Probabilities Entailed by Hardy-Weinberg Equilibrium 487
VII. Discussion and Summary 488
Acknowledgments 490
References 490
Chapter 17: Study Designs for Genome-Wide Association Studies 493
I. Introduction 494
II. Basic Principles of Association Design 499
A. Retrospective case-control studies 500
B. Prospective cohort and nested case-control studies 504
C. Continuous phenotypes 505
D. Using genome-wide SNP data to adjust for population stratification and other biases 506
E. Family-based designs 507
III. The Genetic Architecture of Complex Traits 508
IV. Genotyping Technologies 510
V. Power Calculations for Multistage Design 513
A. Appropriate significance thresholds for power calculations 513
B. Power and cost calculations for multistage designs 515
VI. Discussion 521
References 523
Chapter 18: Ethical, Legal, and Social Implications of Biobanks for Genetics Research 533
I. Introduction 534
II. Governance of Biobanks 535
III. Risks and Benefits 540
A. Individual risks and potential benefits 541
B. Societal risks and benefits 542
IV. Recruitment of Vulnerable and Minority Populations 544
V. Informed Consent 546
A. Approaches to biobank consents 547
B. Tiered consent 548
C. General consent 549
VI. Storage of Genetic Information„Optimizing Privacy 550
VII. Biospecimen/Data Access 552
VIII. Ownership and Intellectual Property 553
IX. Disclosure of Research Results 555
A. Individual research results 555
B. Aggregate research results 562
X. Commercialization of Biobanks 564
XI. Conclusion 566
References 567
Part IV: Promising Topics 573
Chapter 19: Admixture Mapping and the Role of Population Structure for Localizing Disease Genes 575
I. Introduction 576
II. Population Genetic Structure in Humans 577
III. Population Admixture 578
IV. Methods of Admixture Mapping 579
A. The basic idea 579
B. Test statistics 581
C. Inferring locus-specific ancestry 585
D. Genome-wide significance level 586
E. Software available 588
V. Design Consideration 589
A. Markers for admixture mapping 589
B. Power and sample size 590
VI. Prospects and Concerns of Admixture Mapping 591
VII. Conclusions 593
Acknowledgment 594
References 594
Chapter 20: Integrating Global Gene Expression Analysis and Genetics 599
I. Introduction 600
II. Technical and Experimental Design Issues for Microarrays 603
A. DNA microarray platforms 604
B. Microarray data analysis 609
III. Identifying Disease Candidates Using DNA Microarrays 610
A. Gene expression catalogs 610
B. Differential gene expression in disease 611
C. Functional annotation of gene expression patterns 612
D. Identification of disease biomarkers 614
IV. Integration of Genetics and Genomics 614
A. Mapping gene expression QTL 616
B. Prioritizing candidate genes 618
C. Modeling causal interactions 620
D. Gene coexpression networks 620
E. Genetical genomics in human studies 622
V. Conclusions 624
References 625
Chapter 21: A Systems Biology Approach to Drug Discovery 631
I. Introduction 632
II. Causal Inference: An Integrated Approach to Drug Discovery 634
III. Coexpression Networks 640
A. Constructing weighted and unweighted coexpression networks 632
B. Using genetics in constructing coexpression networks 642
C. Identifying modules of highly interconnected genes in coexpression networks 643
IV. Probabilistic Causal Networks: Bayesian Networks as a Framework for Data Integration 645
A. Bayesian networks 646
B. Deriving structure priors from genetic data 647
C. Structure priors derived from other data sources 650
V. An Insightful Example: Integrating Data Leads to the Identification of Gene Affecting Plasma Cholesterol Levels 653
VI. Conclusions 632
References 658
Chapter 22: The Promise of Composite Likelihood Methods for Addressing Computationally Intensive Challenges 665
I. Introduction 666
II. Composite Likelihood Methods 666
III. Applications in Population Genetics 669
IV. Applications in Fine Mapping of Disease Mutations 671
A. Terwilliger's method 672
B. Devlin et al.'s method 673
C. Maleacutecot model for linkage disequilibrium 674
D. Other methods 676
V. Other Applications 677
VI. Prospectives and Discussion 678
References 680
Chapter 23: Comparative Genomics for Detecting Human Disease Genes 683
I. Introduction 684
II. The Power and Promise of Comparative Genomics 687
A. Characterization of genes and their regulation 688
III. Animal Models for Human Disease 693
A. QTL to gene 694
B. Sequence to function 699
C. Comparing phenomes 702
IV. Comparative Genomics and Building Better Animal Models 704
A. Transgenesis and mutagenesis 704
B. Humanizing disease models 706
C. Comparative QTL 707
D. Genes positionally cloned 708
V. Discussion 709
References 711
Part V: Outstanding Challenges 727
Chapter 24: From Genetics to Mechanism of Disease Liability 729
I. Introduction 730
II. Identification of QTL by Statistical Analysis 731
A. QTL mapping in human genetics 731
B. Mapping determinants of essential hypertension in humans 731
C. The predominant role of the kidney in EH 732
D. From QTL to genes in rodents 733
E. Few genes, if any, explaining a blood pressure QTL have been identified by the traditional approach 734
III. A Strong QTL For Sodium Sensitivity in the C57BL/6J Mouse Inbred Line 735
A. QTL mapping of NaS between C57BL/6J and A/J 735
B. Combining data for the chromosome 4 QTL 736
C. A QTL that accounts for a small proportion of the total variance of a quantitative phenotype of low specificity, such as BP, may not be amenable to fine mapping and positional cloning 736
IV. From QTL to Genes: Gene Prioritization 741
A. Gene prioritization based on differential expression in select tissues or cells 742
B. Gene prioritization based on in silico search of functional annotations and genomic differences among strains 744
V. From QTL to Genes: High-Throughput Functional Screens 749
A. The ultimate need for a functional proof 749
B. The power of high-throughput functional screens 749
C. Functional analysis of sodium transport in epithelial cells of the kidney 750
VI. Perspective 751
References 751
Chapter 25: Into the Post-HapMap Era 755
I. Introduction 756
II. Linkage Mapping and Cytogenetic Assignment 757
III. Association Mapping and Functional Assignment 758
IV. Fully Parametric Analysis 760
V. Meta-Analysis 760
VI. Analysis of msSNPs 761
VII. The False Discovery Rate 762
VIII. Early-HapMap Projects 763
IX. Genome-Wide Association Scans 765
X. What Next? 766
References 768
Index 771