Strategies in Biomedical Data Science (eBook)

JAY A. ETCHINGS is the director of operations at Arizona State University's Research Computing program, where he is responsible for developing innovative architectures to progress fluid technical environments supporting highly computational workloads, peta-scale data analysis, next-generation cyber capabilities, and emerging network innovations.

Foreword xi

Acknowledgments xv

Introduction 1

Who Should Read This Book? 3

What's in This Book? 4

How to Contact Us 6

Chapter 1 Healthcare, History, and Heartbreak 7

Top Issues in Healthcare 9

Data Management 16

Biosimilars, Drug Pricing, and Pharmaceutical Compounding 18

Promising Areas of Innovation 19

Conclusion 25

Notes 25

Chapter 2 Genome Sequencing: Know Thyself, One Base Pair at a Time 27

Content contributed by Sheetal Shetty and Jacob Brill

Challenges of Genomic Analysis 29

The Language of Life 30

A Brief History of DNA Sequencing 31

DNA Sequencing and the Human Genome Project 35

Select Tools for Genomic Analysis 38

Conclusion 47

Notes 48

Chapter 3 Data Management 53

Content contributed by Joe Arnold

Bits about Data 54

Data Types 56

Data Security and Compliance 59

Data Storage 66

SwiftStack 70

OpenStack Swift Architecture 78

Conclusion 94

Notes 94

Chapter 4 Designing a Data-Ready Network Infrastructure 105

Research Networks: A Primer 108

ESnet at 30: Evolving toward Exascale and Raising Expectations 109

Internet2 Innovation Platform 111

Advances in Networking 113

InfiniBand and Microsecond Latency 114

The Future of High-Performance Fabrics 117

Network Function Virtualization 119

Software-Defined Networking 121

OpenDaylight 122

Conclusion 157

Notes 157

Chapter 5 Data-Intensive Compute Infrastructures 163

Content contributed by Dijiang Huang, Yuli Deng, Jay Etchings, Zhiyuan Ma, and Guangchun Luo

Big Data Applications in Health Informatics 166

Sources of Big Data in Health Informatics 168

Infrastructure for Big Data Analytics 171

Fundamental System Properties 186

GPU-Accelerated Computing and Biomedical Informatics 187

Conclusion 190

Notes 191

Chapter 6 Cloud Computing and Emerging Architectures 211

Cloud Basics 213

Challenges Facing Cloud Computing Applications in Biomedicine 215

Hybrid Campus Clouds 216

Research as a Service 217

Federated Access Web Portals 219

Cluster Homogeneity 220

Emerging Architectures (Zeta Architecture) 221

Conclusion 229

Notes 229

Chapter 7 Data Science 235

NoSQL Approaches to Biomedical Data Science 237

Using Splunk for Data Analytics 244

Statistical Analysis of Genomic Data with Hadoop 250

Extracting and Transforming Genomic Data 253

Processing eQTL Data 256

Generating Master SNP Files for Cases and Controls 259

Generating Gene Expression Files for Cases and Controls 260

Cleaning Raw Data Using MapReduce 261

Transpose Data Using Python 263

Statistical Analysis Using Spark 264

Hive Tables with Partitions 268

Conclusion 270

Notes 270

Appendix: A Brief Statistics Primer 290

Content Contributed by Daniel Peñaherrera

Chapter 8 Next-Generation Cyberinfrastructures 307

Next-Generation Cyber Capability 308

NGCC Design and Infrastructure 310

Conclusion 327

Note 330

Conclusion 335

Appendix A The Research Data Management Survey: From Concepts to Practice 337

Brandon Mikkelsen and Jay Etchings

Appendix B Central IT and Research Support 353

Gregory D. Palmer

Appendix C HPC Working Example: Using Parallelization Programs Such as GNU Parallel and OpenMP with Serial

Tools 377

Appendix D HPC and Hadoop: Bridging HPC to Hadoop 385

Appendix E Bioinformatics + Docker: Simplifying Bioinformatics Tools Delivery with Docker Containers 391

Glossary 399

About the Author 419

About the Contributors 421

Index 427