Single Cell Methods to Learn Transcription Factor Interactions and Necessary Noncoding DNA During Zebrafish Somitogenesis

Abstract

The precise control of gene expression during embryonic development is orchestrated by complex networks of transcription factors (TFs) and their interactions with noncoding regulatory DNA elements. However, our understanding of how these elements function in vivo and how TFs cooperate to drive cell fate decisions remains incomplete and limited in scope. In this thesis, I will present a method sciPlex ATAC-seq, a multiplexed low cost method for single-cell assay for transposable accessible chromatin. Pairing this method with sciPlex RNA-seq, a similar method for transcriptomic measurement, allowed me to build an integrated single-cell atlas of zebrafish embryogenesis and organogenesis from unpaired data to map cis-regulatory elements during zebrafish development. This approach enables single-cell resolution mapping of accessible regulatory DNA in hundreds to thousands of individually indexed embryos, allowing the identification of enhancers and TF binding motifs across hundreds of cell types. I further integrate these data with deep learning models to infer TF-TF cooperativity and test mechanistic predictions using F0 CRISPR injected zebrafish embryos. Following up on novel interactions from our single-cell perturbation experiments, I developed and applied methods to identify TF perturbation responsive noncoding DNA elements and tested their sufficiency and necessity during zebrafish somitogenesis. This framework offers a low-cost and scalable approach to decode the regulatory logic of development and provides a blueprint for functionally annotating the noncoding genomes of multicellular models.