MS, University of Cincinnati, 2024, Medicine: Genetic Counseling

Copy number variant (CNV) analysis detects structural genomic alterations, including microdeletions and microduplications. Despite the implication of CNVs in various human diseases, there has not been consistent pathogenicity classification of CNVs. To address this issue, the ACMG and the NIH-funded Clinical Genome Resource (ClinGen) published joint technical standards in 2020 (Riggs et al. 2020). They also developed a supplemental calculator tool that can be used to quickly tabulate CNV pathogenicity based on their recommendations. Since publication, there has not been an examination of the technical standards and how they have affected laboratory practices. We surveyed over 300 personnel at U.S. clinical genetics laboratories to evaluate their use of the ACMG/ClinGen technical standards. We also explored how the standards affected CNV interpretation and reporting. Data were summarized as frequencies and percentages. Of 55 respondents, 42 (74%) reported that they use the technical standards. Among users of the CNV standards, 30 respondents (71%) reported that they use the standards for primary interpretation, 22 (52%) for resolving conflicting interpretations, and 12 (29%) for secondary confirmation. Additionally, 11/42 (26%) responded that the number of reported CNVs per report has changed within their laboratory since the implementation of the standards. All 11 reported an increase in the report of variants of unknown significance (VUS), two reported an increase in the report of likely pathogenic (LP) variants, and two reported an increase in the report of likely benign (LB) variants. Other respondents reported decreases in pathogenic (n=3), LP (n=3), LB (n=4), and benign (n=4) CNVs reported. Our findings indicate most laboratory personnel use the technical standards for CNV analysis, particularly for primary interpretation. However, only 26% (11/42) of respondents felt use of the standards changed CNV reporting, with an increase in (open full item for complete abstract)

Committee: Melanie Myers Ph.D. (Committee Chair); Stephanie Balow (Committee Member); Valentina Pilipenko Ph.D. (Committee Member); Leandra Tolusso M.S. (Committee Member); Teresa Smolarek Ph.D. (Committee Member) Subjects: Genetics

MS, University of Cincinnati, 2022, Medicine: Genetic Counseling

Genome-wide assessment of copy number variants (CNVs) by chromosomal microarray (CMA) is a first-line test offered to individuals with developmental delay, multiple congenital anomalies, autism spectrum disorder, and other indications. Exome reanalysis studies have demonstrated that reanalysis of genetic tests over time can yield additional diagnoses due to updates in genetic databases and the literature. Compared to exome reanalysis studies, there is limited data regarding the yield of microarray reanalysis. We conducted a manual reanalysis of 154 non-diagnostic patient SNP microarrays five years after they were initially performed in 2016. We found that 25/339 (7.4%) CNVs changed in classification after reanalysis with 24/154 (15.6%) patients having at least one variant that changed in classification. Of those variants that changed, 15/25 changed from benign to uncertain significance. However, only 2/154 patients (1.3%) had a diagnostic change, with a variant upgraded to likely pathogenic or pathogenic. The most common reasons variants changed in classification were changes in CNV classification standards, in particular the decoupling of patient phenotype to variant classification and new scoring guidelines. Overall, we demonstrate that while variants can be classified differently over time, SNP microarray reanalysis yields lower diagnostic rates compared to previously published exome reanalysis studies, if done within a relatively recent time-period of initial analysis (5-6 years).

Committee: Melanie Myers Ph.D. (Committee Member); Alyxis Coyan M.S. (Committee Member); Qiaoning Guan (Committee Member); Wenying Zhang M.D. Ph.D. M.B.A. (Committee Member); Teresa Smolarek Ph.D. (Committee Member); Elizabeth Ulm M.S. (Committee Member); Jie Liu (Committee Member) Subjects: Genetics

PhD, University of Cincinnati, 2021, Engineering and Applied Science: Computer Science and Engineering

We understand the effect of a small fraction of known human genetic variants. Computational approaches can help provide insight into potentially damaging variant effects. One such approach is the classification of damaging and tolerated variants based on functional or structural features. Many existing prediction approaches have been developed to predict this in a meaningful way. However, it is not clear how the performance of these methods varies across diseases, or they depend on the types of functional and structural features used. Therefore, it is essential to analyze disease-specific performance and determine whether these approaches are helpful for a given disease. This research project systematically compares various prediction tools and techniques, relates their disease type performance, and demonstrates the varied performance of existing tools across diseases and biological systems. Next, we design a bioinformatics framework, Variant-Mine, to model and predict damaging variants for a specified disease. The Variant-Mine framework models known pathogenic disease-specific variants based on their functional, structural, or conservation features and identifies similar variants in candidate genes that may have deleterious or modifying effects. To model these effects, we developed a suitable and scalable classification algorithm called Bicluster-based Rule Mining or BBRM, specializing in finding dense enriched regions in noisy and imbalanced data. BBRM is utilized in the Variant-Mine framework to identify feature subspaces within which damaging genetic variants may be more sensitively detected. Lastly, we use the Variant-Mine framework to successfully identify putative damaging variants and their corresponding features in a Cardiomyopathy disease cohort.

Committee: Bruce Aronow Ph.D. (Committee Chair); Raj Bhatnagar Ph.D. (Committee Member); Ali Minai (Committee Member); Jaroslaw Meller Ph.D. (Committee Member); Lisa Martin Ph.D. (Committee Member) Subjects: Computer Science

Master of Science, The Ohio State University, 2017, Genetic Counseling

Introduction: Genetic testing for cardiovascular disease (CVD) is a powerful tool that enables clinicians to identify genetic forms of CVD and predict the risk for CVD in at-risk family members. Cardiovascular genetic testing has advanced over the past ten years, but these advancements have posed new challenges mainly in the field of variant classification. To address these challenges, the American College of Medical Genetics and Genomics (ACMG) published guidelines for the interpretation of sequence variant interpretation in 2015. Goal: The goal of this study was to determine what impact the ACMG guidelines have on variant classification in clinical cardiovascular genetics. Methods: We performed a retrospective chart review to identify patients who underwent clinical genetic testing and were found to have a variant identified in a gene associated with CVD. For each variant, systematic evidence review was performed by collecting information from both public and private variant databases and PubMed. We applied the ACMG guidelines to each variant for classification, which were compared to classifications provided on patients' genetic test reports. Results: This study identified 223 unique variants in 237 patients. Eighty (36%) of the variants resulted in classifications that differed from their clinical reports. Twenty-seven (34%) of these reclassifications were determined to be clinically significant. In total, these variant classifications affected 101 patients in a single clinical setting. For 39 patients (39% of 101), these reclassifications would result in changes in medical management recommendations for their at-risk relatives. Conclusion: Application of the ACMG guidelines resulted in a change of classification for approximately one-third of the variants in this study. Clinical genetic counselors can have a more active role in the process of variant classification. It is important for variant classifications to be updated over time (open full item for complete abstract)

Committee: Leigha Senter-Jamieson MS, LGC (Advisor); Amy Sturm MS, LGC (Committee Member); Robert Pyatt PhD (Committee Member); Sayaka Hashimoto MS, LGC (Committee Member); Ana Morales MS, LGC (Committee Member) Subjects: Genetics