Clinical Genomics

Shashikant Kulkarni, Somak Roy

• Publisher: Academic Press
• Year: 2014
• ISBN:
  9780124051737
• Topic: Biological Sciences
• Subtopic: Genetics & Genomics
• Index: Biological Sciences

Clinical Genomics provides an overview of the various next-generation sequencing (NGS) technologies that are currently used in clinical diagnostic laboratories. It presents key bioinformatic challenges and the solutions that must be addressed by clinical genomicists and genomic pathologists, such as specific pipelines for identification of the full range of variants that are clinically important.

This book is also focused on the challenges of diagnostic interpretation of NGS results in a clinical setting. Its final sections are devoted to the emerging regulatory issues that will govern clinical use of NGS, and reimbursement paradigms that will affect the way in which laboratory professionals get paid for the testing.

• Simplifies complexities of NGS technologies for rapid education of clinical genomicists and genomic pathologists towards genomic medicine paradigm
• Tried and tested practice-based analysis for precision diagnosis and treatment plans
• Specific pipelines and meta-analysis for full range of clinically important variants

Section III

Interpretation

Outline

Chapter 12 Reference Databases for Disease Associations

Chapter 13 Reporting of Clinical Genomics Test Results

Chapter 14 Reporting Software

Chapter 15 Constitutional Diseases

Chapter 16 Targeted Hybrid Capture for Inherited Disease Panels

Chapter 17 Constitutional Disorders

Chapter 18 Somatic Diseases (Cancer)

Chapter 19 Targeted Hybrid-Capture for Somatic Mutation Detection in the Clinic

Chapter 20 Somatic Diseases (Cancer)

Chapter 12

Reference Databases for Disease Associations

Wendy S. Rubinstein¹, Deanna M. Church² and Donna R. Maglott¹, ¹National Center for Biotechnology Information/National Library of Medicine/National Institutes of Health, Bethesda, MD, USA, ²Personalis, Inc., Menlo Park, CA, USA

Making sense of the human variation in the context of human disease has many challenges. A medical genetics professional faced with a list of identified variants must be able to evaluate, and reevaluate, whether there are prior reports about any variant, what the current interpretation of that variant might be, and/or if there are predicted consequences. The sources of publicly available information are diverse, so the genetics professional needs to know not only how to find information, but also how large centralized databases and datasets are maintained in case a current interpretation is not highly supported. This chapter provides an overview of how human genetic information is ascertained—both the common variation to provide the baseline of information about variation unlikely to be the primary cause of a disorder, and the rarer variation that may contribute to disease. It summarizes the features of major databases that archive submissions about variation (dbSNP, dbVar), about variation–disease associations (dbGaP), and clinical interpretations of variation (ClinVar). It also summarizes major tools that support access to these data by several paths, e.g., by location on the genome, by gene, and by disease or measured phenotype. The chapter also describes community efforts to encourage data sharing and to develop standards that promote the development of clinical grade databases.

Keywords

Sequence variation; ClinVar; dbGaP; dbSNP; dbVar; MedGen; phenotype; disease; pathogenicity; web sites; GWAS; ClinGen

Outline

Introduction 192

Identification and Validation of Human Variation 193

Methods for Identifying Human Variation 193

Targeting Known Sequences 193

Sequence Discovery 193

Sequencing and Identifying Differences Relative to a Reference 194

Understanding a Reference Assembly 194

Validation of Variant Calls 195

Identification of Common Variation 195

Overview 195

HapMap 196

1000 Genomes Project 197

NHLBI-ESP 198

Interpretation of Common Variation 199

Determining Association Between Phenotype and Common Variation 199

Health-Related Phenotypes (GWAS) 199

Molecular Phenotypes (Gene Expression/GTex) 199

Inferring Lack of Causality Based on Allele Frequency 199

Defining Diseases and Phenotypes 199

UMLS and MedGen 200

OMIM® 200

Orphanet 200

Human Phenotype Ontology 201

Combining Phenotypic Features 201

Representation of Variation Data in Public Databases 201

Archives of Variants 201

dbSNP 201

dbVar 202

Archives of Variant/Phenotype Relationships 202

dbGaP 202

ClinVar 204

OMIM 204

DECIPHER 204

COSMIC 204

PubMed/PubMedCentral 205

Data Access and Interpretation 205

By One or More Genomic Locations 205

By Gene 208

By Condition or Reported Phenotype 208

By Attributes of a Particular Variant 208

Variants in the ACMG Incidental Findings Gene List 209

Determination of Variant Pathogenicity 210

Expert Panels and Professional Guidelines 210

ClinGen: A Centralized Resource of Clinically Annotated Genes and Variants 211

Standard Setting for Clinical Grade Databases 211

Global Data Sharing 212

GA4GH and the Beacon Project 212

Conclusion 213

References 213

List of Acronyms and Abbreviations 215

Key Concepts

• The approach of testing specific genes and variants based on pre-established evidence of causation is shifting toward the use of NGS techniques that can provide sensitivity at lower cost. However, the detection of many novel sequence variants requires automated approaches to facilitate interpretation.

• Interpretation of the clinical significance of sequence variation requires evaluation of the relevant evidence, but such evidence may be difficult to gather or altogether lacking. Centralized databases such as those at National Center for Biotechnology Information (NCBI) support many reference databases that are designed for representing variation and for the reporting of relationships to phenotypes. The major databases that archive submissions include Database of Short Genetic Variations (dbSNP) and Database of Structural Variation (dbVar) (for small and large variation, respectively), Database of Genotypes and Phenotypes (dbGaP) (for variation–disease associations), and Database of Clinical Variation (ClinVar) (for clinical interpretations of variation). Medical Genetics resource at NCBI (MedGen) harmonizes phenotype terminologies and supports computational access to phenotype data.

• Interpretation of variants requires understanding of the role of the human reference assembly and its annotation, which are periodically versioned based on new knowledge. Professionals using NGS techniques need to be aware of problematic regions within the human reference assembly.

• Large-scale efforts such as HapMap, the 1000 Genomes Project, and NHLBI-ESP, that characterize genetic variation in diverse population groups, support assessments of the frequency of human variation. ClinVar, dbSNP, and Variation Viewer support searching and filtering functions related to allele frequencies. dbSNP provides variant call file (VCF) files of common variants not known to be disease-related.

• Laboratories that utilize NGS technology to detect sequence variation encounter variants that have not previously been reported, and proprietary databases may be inadequate for the reliable interpretation of clinical significance. Consequently, there is increasing participation in public data sharing, and several community efforts are under way to develop standards that promote the development of clinical grade databases.

• Professional societies, laboratories, accrediting organizations, and federal agencies have been developing professional guidelines and quality assurance programs that support the application of NGS technology in the clinical realm. The GeT-RM NGS dataset and browser, as well as the Genome In A Bottle project, support assessment of the analytical validity of any variant call.

Introduction

Medical professionals must increasingly evaluate genetic variant data generated by clinical laboratories, research studies, and direct-to-consumer testing obtained by their patients. Not only is there a growing number of conditions for which testing is available, but also the previous paradigm of focused testing for genes with well-accepted clinical consequences, and for specific variants in those genes, has shifted toward the use of massively parallel sequencing to identify common and rare variants across many or all genes in the genome. Traditional single gene tests have begun to be replaced by complex panels that assay multiple genes with varying levels of evidence for disease association and limited characterization of the variation spectrum. Also, patients with presumably genetic, undiagnosed diseases are increasingly undergoing whole exome and whole genome analyses using next generation sequencing (NGS) methods. Not only does this pose challenges to the evaluation of variants in genes with a plausible contribution to the phenotype, but variants can also be uncovered in genes with unexpected but important clinical consequences.

Technical advances in sequencing coupled with reduced cost have led to an abundance of genetic variant data across the clinical realm, with the promise of more to come. The interpretation of genetic variant data has become a rate-limiting step in the utilization of this information. Factors that make interpretation of variant data rate limiting include lack of clinical standards for interpreting primary NGS results, inconsistencies in reporting clinical variants in commonly used file formats, limited adoption of variant nomenclature standards, information gaps between those capturing variant data and those evaluating the phenotypes of the test subjects, lack of standard interoperable phenotype vocabularies, the relative immaturity of resources containing population-specific variant frequency data, insufficient cross-training between the research-based groups practiced in managing big data and the clinical community, and limited data sharing which slows well-powered analyses. Perhaps the single most important barrier to the interpretation of variant data is the...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Dedication
• List of Contributors
• Foreword
• Preface
• Acknowledgments
• Section I: Methods
• Section II: Bioinformatics
• Section III: Interpretation
• Section IV: Regulation, Reimbursement, and Legal Issues
• Index