Linking GWAS variants to function with single-cell pooled CRISPR screens

PROJECT SUMMARY/ABSTRACT Genome-wide association studies (GWAS) have identified thousands of common and rare genetic variants associated with complex traits and common diseases. Most variants map to the 98% of the genome that is noncoding, with their target genes or function largely unknown. This is the variant-to-function problem (V2F), and

solving it remains a major hurdle in human genetics research. To help solve V2F, I propose to develop modular workflows combining GWAS variant prioritization methods and pooled single-cell CRISPR screens for target gene identification. I have developed an integrative approach combining highly polygenic blood trait GWASs and

pooled single-cell CRISPR inhibition (CRISPRi) screens in a human erythroid progenitor cell model (K562), to identify target genes: Systematic Targeting and Inhibition of Noncoding GWAS loci with single-cell sequencing (STING-seq). STING-seq can functionally dissect multiple GWAS loci in a massively parallel fashion, identifying

target genes in cis as well as trans-regulatory networks. Here, I will develop STING-seq further and examine its generalizability for other GWAS traits and their cell models. First, I will expand STING-seq with precise variant insertion, developing base editing STING-seq (Bee-STING) for high-throughput measurements of GWAS variant

effects on target genes and regulatory networks. Second, I will develop modular workflows for GWAS variant prioritization for STING-seq, targeting sets of variants with distinct selection criteria to increase STING-seq’s target gene and regulatory network discovery rate. Third, I will focus STING-seq on new GWAS traits and cell

models to examine its generalizability, first piloting STING-seq for another highly polygenic complex trait, bone mineral density, with a human osteoblast cell model (hFOB). In the long-term, these aims will help solve V2F for human genetics research, as their continued development and application will improve our understanding of how

GWAS variants causally influence complex traits and common diseases. I have a comprehensive training plan in place with my primary mentors, Dr. Neville Sanjana (genome engineering) and Dr. Tuuli Lappalainen (gene regulation), my mentorship committee members, Dr. David Knowles (machine learning), Dr. Aravinda

Chakravarti (human genetics), Dr. Charles Farber (bone biology), and my collaborator Dr. Eugene Katsevich (statistical methods). This plan will continue my training in dissecting GWAS variant function with multiple computational and experimental approaches, along with additional training in grant writing, mentoring students,

teaching courses, and presenting at research conferences. The full mentorship committee will direct me to pertinent literature, offer advice on my research program, and provide guidance as I navigate the academic job market. The New York Genome Center is the ideal training location for me, given its cutting-edge facilities,

plentiful opportunities for career and intellectual development, and collaborative research environment. Upon completion of this training program, I will be well-positioned to lead my own interdisciplinary research lab and become a leader in the fields of human complex traits genetics and genome engineering.