Data Mining and Applications in Genomics (eBook)

Preface 7
Contents 9
Introduction 13
1.1 Data Mining Algorithms 13
1.2 Advances in Genomic Techniques 16
1.3 Case Studies: Building Data Mining Algorithms for Genomic Applications 22
Data Mining Algorithms 26
2.1 Dimension Reduction and Transformation Algorithms 26
2.2 Machine Learning Algorithms 31
2.3 Clustering Algorithms 41
2.4 Graph Algorithms 44
2.5 Numerical Optimization Algorithms 46
Advances in Genomic Experiment Techniques 50
3.1 Single Nucleotide Polymorphisms (SNPs) 50
3.2 HapMap Project for Genomic Studies 53
3.3 Haplotypes and Haplotype Blocks 56
3.4 Genomic Analysis with Microarray Experiments 60
Case Study I: Hierarchical Clustering and Graph Algorithms for Tag- SNP Selection 64
4.1 Background 64
4.2 CLUSTAG: Its Theory 69
4.3 Experimental Results of CLUSTAG 73
4.4 WCLUSTAG: Its Theory and Application for Functional and Linkage Disequilibrium Information 78
4.5 WCLUSTAG Experimental Genomic Results 81
4.6 Result Discussions 82
Case Study II: Constrained Unidimensional Scaling for Linkage Disequilibrium Maps 83
5.1 Background 83
5.2 Theoretical Background for Non-parametric LD Maps 90
5.3 Applications of Non-parametric LD Maps in Genomics 94
5.4 Developing of Alterative Approach with Iterative Algorithms 106
5.5 Remarks and Discussions 124
Case Study III: Hybrid PCA-NN Algorithms for Continuous Microarray Time Series 126
6.1 Background 126
6.2 Motivations for the Hybrid PCA-NN Algorithms 130
6.3 Data Description of Microarray Time Series Datasets 131
6.4 Methods and Results 132
6.5 Analysis on the Network Structure and the Out- of- Sample Validations 136
6.6 Result Discussions 138
Discussions and Future Data Mining Projects 139
7.1 Tag-SNP Selection and Future Projects 139
7.2 Algorithms for Non-parametric LD Maps Constructions 144
7.3 Hybrid Models for Continuous Microarray Time Series Analysis and Future Projects 146
Bibliography 148