Ph.D. candidate at Johns Hopkins University

Research

Current Research

PIWI-interacting RNAs (piRNAs) are expressed in a sex-specific manner across many species, including human. Understanding how sex-specific piRNA expression regulates germline gene expression is paramount to advancing our knowledge of piRNA-mediated regulation beyond transposon silencing and improving therapeutic strategies for infertility. Using C. elegans as a model, my current research seeks to define the regulatory machinery underlying sex-specific piRNA expression and identify endogenous gene targets of piRNAs to provide a deeper understanding of piRNAs in germline development.

My long-term research interests involve using computational approaches to better understand genetic, epigenetic, and other regulatory mechanisms that control gene expression within certain cell types, tissues, and organs. I am also interested in developing computational tools to investigate genomic variations and their functional consequences on gene expression in human health and disease.

Scientific Contributions

Computational methods to detect aneuploidy in single-cell data

In vitro fertilized human embryos show substantial levels of chromosome mis-segregation which leads to aneuploidy, a major cause of pregnancy loss in humans. Aneuploidy arising during mitosis results in mosaic embryos. Prior to our study, estimates of mosaicism in human embryos ranged from 4% to 90%, raising long-standing questions as to the true incidence of mosaicism and its implications in the context of preimplantation genetic testing for aneuploidy. We leveraged single-cell RNA-seq data and developed a method to integrate signatures of gene expression changes and allelic imbalance, improving aneuploidy detection and characterizing patterns of aneuploidy throughout preimplantation development. Specifically, I performed aneuploidy detection based on gene expression. Our work provides an overview of the diverse and widespread nature of mosaic aneuploidies in preimplantation embryos and suggests that low-level mosaicism is a common feature of early human development.

Age-related macular degeneration (AMD) is a complex, neurodegenerative disease that progressively causes irreversible vision loss. Although genome-wide association studies have identified 52 variants associated with AMD susceptibility, mechanisms underlying retina physiology and AMD pathogenesis are poorly understood. We procured post-mortem human retina samples for high-throughput sequencing to (1) determine global gene expression patterns during human retina development, (2) characterize transcriptional changes associated with AMD, and (3) assess expression profiles of genes with rare variants in families with a history of AMD.

In one project to generate a high-resolution transcriptional profile of early retinal development, I extracted RNA from fetal retina and prepared mRNA-seq libraries for high-throughput sequencing, and developed the bioinformatics pipeline for transcriptome analysis. Together with DNase-seq and immunohistochemical studies, we constructed a comprehensive profile of the developing human retina that researchers can now refer to for molecular staging of stem cell-derived retinal organoid cultures to study and develop therapies for degenerative retinal diseases.

In another project we performed a large-scale study using genotyping and mRNA-seq to systematically determine how gene expression patterns differ between those with and without AMD, and whether or not that pattern of expression differs between various AMD phenotypes. By expression quantitative trait locus analysis integrating a recent AMD genome-wide association study, we determined candidate target genes at six reported AMD loci. In addition, we detected three novel AMD risk genes using transcriptome-wide association analysis. I was involved in both the experimental and computational aspects of this long-term project. I helped extract RNA from tissue samples, and together with a post-doctoral fellow, prepared over 500 mRNA-seq libraries for high-throughput sequencing. I developed the overall bioinformatics workflow and performed transcriptome analysis. I used the transcriptome profile I constructed from healthy retina in this large-scale study in another project to assess the expression of candidate genes with rare protein-coding variants in several AMD families. Together, our work is a valuable genomic resource and serves as a basis for identifying mechanisms underlying normal and pathologic functions in human retina.