Dissertation
A data-driven framework for CNV pathogenicity classification
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Autumn 2022
DOI: 10.25820/etd.006768
Abstract
The phenotypic spectrum of human microdeletion and microduplication syndromes (MMS) is heterogeneous but often involves intellectual disability, autism spectrum disorders, dysmorphic features, and/or multiple congenital anomalies. While the common recurrent copy number variants (CNVs) which underlie these MMS have been well-studied (“pathogenic CNVs”), the expansion of clinical genomic testing has led to the identification of many rare non-recurrent CNVs and CNVs in unaffected control populations (“benign CNVs”). To date, hundreds of unique MMS CNVs have been reported, maps of normal copy number variation have been developed, and thousands of variants of unclear clinical significance (VUS) have been identified. VUS represent a large subset of rare and non-recurrent CNVs in which there is insufficient evidence to assign a benign or pathogenic classification.
The interpretation of clinical chromosomal microarrays (CMAs) has historically relied on the relevance of identified CNVs to the clinical phenotype. However, new interpretation guidelines focus on standardizing pathogenicity classifications based on genomic location, gene content, and previous publications rather than the immediate clinical relevance. Discriminating benign CNVs from those that are pathogenic for rare genetic disorders, therefore, relies on understanding what regions of the human genome are tolerant to copy number variation. Further complicating the characterization of CNV pathogenicity are discordant classifications of variants across clinical laboratories, phenotypic heterogeneity, incomplete penetrance, and the lack of understanding of the contributions of individual genes to the clinical phenotype for well-known and well-characterized pathogenic CNVs. Lastly, to account for advances in our understanding of regions of the genome which are tolerant and intolerant to copy number variation, periodic reanalysis is encouraged. Due to the volume of CNVs that are ascertained clinically, reanalysis of pathogenicity classifications is a significant undertaking which requires the aid of computational tools.
In this thesis work described herein, I provide a data-driven framework for improving the pathogenicity classification of clinically ascertained CNVs with the goal of decreasing the number of VUS test results which are returned clinically. The overall project was driven by two main aims: (1) delineate regions of the human genome which are tolerant and intolerant to copy number variation and (2) develop a framework to facilitate the prioritization and re-analysis of CNV pathogenicity on a large-scale. The findings from my dissertation broaden the knowledgebase of both pathogenic and benign CNVs, provide a resource for clinicians in the interpretation of non-recurrent CNVs, and provide two computational tools which can be utilized by the broader research community to continually generate updated maps of benign regions (Benign-Ex) and enable large-scale high-throughput CNV classification, reanalysis, and prioritization (DISCRIMINATOR).
Details
- Title: Subtitle
- A data-driven framework for CNV pathogenicity classification
- Creators
- Alyssa Sue Wetzel
- Contributors
- Benjamin W Darbro (Advisor)Patrick J Breheny (Committee Member)Anne E Kwitek (Committee Member)Jake J Michaelson (Committee Member)Thomas H Wassink (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Genetics
- Date degree season
- Autumn 2022
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.006768
- Number of pages
- xiv, 525 pages
- Copyright
- Copyright 2022 Alyssa Sue Wetzel
- Language
- English
- Description illustrations
- Illustrations, charts, graphs, tables
- Description bibliographic
- Includes bibliographical references (pages 516-525).
- Public Abstract (ETD)
According to a survey conducted by the National Institute of Health, nearly 20% of the US population aged 3-18 are affected by developmental disabilities, and this proportion has increased over time. While a variety of genetic and environmental factors have been implicated, approximately 10-20% of cases can be attributed to genomic copy number variations (CNVs) – sections of DNA that have been duplicated or deleted. For this reason, chromosomal microarrays (CMAs; a clinical genetic test that looks for CNVs) are the first line diagnostic test for individuals with developmental disabilities. The CMA is subject to several limitations: the ability to detect and interpret changes, distinguishing between benign and pathogenic findings, understanding that negative test results do not necessarily rule out a genetic condition, and that individuals with known presumed pathogenic mutations may appear unaffected. While CNVs are found in healthy populations ("benign"), over 150 have been associated with human disease ("pathogenic"), a large proportion of CNVs are classified as variants of unclear clinical significance (VUS; CNVs with insufficient evidence to be classified benign or pathogenic). Of the patients seen locally who undergo CMA testing, nearly 40% receive a VUS as their genetic test result.
Therefore, my thesis project, which is focused on improving the ability to interpret those VUS, has the potential to directly improve the clinical care of patients with a suspected genetic disorder. I developed a comprehensive list of pathogenic CNVs reported in the medical literature, two computational software tools (Benign-Ex and DISCRIMINATOR) and characterized clinically identified CNVs.
- Academic Unit
- Interdisciplinary Graduate Program in Genetics
- Record Identifier
- 9984362857802771
Metrics
4 File views/ downloads
40 Record Views