2022 PCGC/CDDRC Fellows

Andrew Blair photo

Andrew Blair, Ph.D. Candidate

University of California San Francisco 

Andrew Blair is a Biological and Medical Informatics Ph.D. candidate at UC San Francisco and a trainee in the Impact of Genomic Variation on Function (IGVF). His research interest is deciphering the regulatory role of genomic elements in human cardiac development and disease. Before starting his Ph.D., he received an M.S. in Bioinformatics from UC Santa Cruz, supporting the California Institute of Regenerative Medicine’s Heart of Cell team. Following his M.S., he worked as a Genomic Analyst for the Gladstone Cardiovascular Disease Institute and Genentech. His current research with Dr. Nadav Ahituv focuses on building congenital heart disease (CHD) candidate regulatory element models for experimental characterization and functionalization.


Multi-omic Modeling of Candidate Regulatory Elements to Decipher Congenital Heart Disease

Variants in gene regulatory elements are a significant cause of congenital heart disease (CHD). However, associating variants in candidate regulatory elements (CREs) to CHD is challenging to identify and characterize. Using the CDDRC multi-omics and PCGC dbGAP/TOPMeD data in the BioData Catalyst ecosystem, along with an ensemble of deep learning model predictions, I will develop spatiotemporal networks for cardiac cell-type-specific CREs. This network will allow users to link CHD-associated CREs to target gene regulatory domains, including which time point and in vitro system or organism to best model CHD. Using predictions prioritized by the network, over 50,000 CHD-associated CREs will be functionally evaluated via massively parallel reporter assays (MPRA) in cardiac progenitors, primitive cardiomyocytes, and atrial and ventricular cardiomyocytes, allowing me to improve the networks’ predictions. This work will provide a user, data-driven “report card” from our catalog of functionally characterized CHD-associated CREs, supporting the design of future functional testing and providing a framework to refine our understanding of CHD further.

Leroy Bondhus photo

Leroy Bondhus, Ph.D. Candidate

Human Genetics program at the University of California Los Angeles

Leroy Bondhus is a PhD candidate in the Human Genetics program at the University of California  Los Angeles in the lab of Dr. Valerie Arboleda studying rare chromatin modifier syndromes. His research focuses on developing novel modeling strategies to better understand the epigenetic and transcriptomic changes that occur in normal development and disease. Previous projects he has published on in this space include development of the method and associated R package DMRscaler for identifying regions of differential methylation across the full range of genomic scale from genome wide methylation data, and developing a robust and general procedure for incorporating sample similarity information into measures of gene specificity for quantitatively defining the level of specificity of gene expression to particular biological contexts, such as tissues or cell types. One of his current interests is in developing models of heart development integrating genetic, epigenetic, and phenotypic dimensions of information to enable deeper exploration of the factors that contribute to healthy and disordered heart development. He is enthusiastic about open research initiatives that aim to make biomedical knowledge and the tools for uncovering that knowledge more widely accessible and transparent.


Modelling the developmental and molecular context of CHD associated genetic variation

Congenital heart disease (CHD) describes a wide range of lesions that arise during heart development and adversely affect heart function. Methods and technologies for identifying genetic variation associated with CHD have evolved from early gene mapping work in familial cases of CHD to modern studies identifying de novo and non-coding causes of CHD through exome and whole genome sequencing. A major gap remains, however, in methods for identifying how genetic variants perturb molecular phenotypes during development ultimately resulting in CHD phenotypes. To address this, we will leverage datasets available through the CDDRC, GEO, and elsewhere to develop a scale and relation-aware  multi-omic model of features that characterize heart development and identify specific features of heart development perturbed by variants contributing to CHD identified in the PCGC datasets.

Chani Jo Hodonsky, Ph.D., MPH

University of Virginia Center for Public Health Genomics

Chani Hodonsky is a geneticist and an epidemiologist working with bioinformatic, molecular, and next-gen sequencing technologies to identify genes and genetic variants associated with coronary artery disease. She is committed to improving inclusive study design as well as statistical genetics methods development that will benefit global populations rather than risk increasing current health inequities in the US and abroad. She also works with the Leducq Plaqomics consortium to identify and characterize the presence of somatic clones in atherosclerotic plaque progression. Outside of work she plays water polo, grows hot peppers, and takes way too many pictures of her dog.


Genetic classification of single-ventricle disease subtypes to identify gene expression profile changes in post-Fontan sequelae

Congenital heart defects with single-ventricle physiology (single-ventricle diseases [SVDs]) are polygenic and present heterogeneously. Long-term complications are common, yet early detection and treatment options remain limited. Characterizing distinct SVD subtypes will benefit clinical management and therapeutic development. I propose to use multivariate gene-based testing of SVD endophenotypes in PCGC participants to identify genetic signatures for subtypes within current disease definitions. I will then use deconvolute bulk RNA-sequencing data from CDDRC patients and mouse models to characterize subtype-specific gene expression profiles and evaluate cell-type proportion differences between subtypes. This combined genetic and transcriptomic approach will provide candidates for early-detection biomarkers of clinical manifestations, as well as distinguishing subtype-specific molecular pathways from pleiotropic contributors to the broader SVD phenotypic spectrum.

Jonathan Klonowski, Ph.D. Candidate

Cecilia W. Lo Laboratory, University of Pittsburgh School of Medicine

Jonathan Klonowski is the son of Polish immigrants who grew up on the west side of Chicago and obtained his B.S. from the University of Illinois at Chicago. As a scientific nomad, Jonathan has received training in many biological disciplines including biochemistry, molecular biology, cell biology, virology and developmental biology. Pursuing his PhD at the University of Pittsburgh, School of Medicine’s Integrative Systems Biology program, Jonathan is currently utilizing a combination of genetics, computational biology and bioinformatics to understand the genetic etiology of congenital heart disease.


Role of Nonsense-Mediated mRNA Decay (NMD) Escaping Variants in the Pathogenesis of Congenital Heart Disease.

Premature termination codon (PTC) causing mutations represent a large fraction of clinically relevant pathogenic genomic variation. Typically, PTCs generate transcripts degraded by nonsense-mediated mRNA decay (NMD), causing loss-of-function (LOF) allele. However, PTC containing transcripts can escape NMD, possibly resulting in dominant negative or gain-of-function (DNGOF).  Ability to systematically identify PTC-causing variants predicted to escape NMD will make it possible to investigate the potential role of DNGOF variants in human disease. Herein, we will develop Dockstore software for annotating sequencing data at scale, implementing established and experimentally validated rules for NMD escape. Using this, we will recover predicted NMD escaping variants (pNEVs) among Pediatric Cardiac Genomics Consortium (PCGC) congenital heart disease (CHD) patients to assess their role in CHD pathogenesis. Functional impact will be discerned by cross-species analysis of pNEVs from mutations recovered from mouse mutagenesis screens.