UC San Diego

The epigenome controls the spatial and temporal specificity of gene expression in mammalian cells, guiding different cell fates and developmental trajectories. Our current understanding of how the epigenome regulates transcription is primarily based on bulk epigenetic assays. While these assays provide valuable information, the data represents population averages, which obscure detailed cell type or cell state-specific information in heterogeneous systems. Moreover, bulk data is dominated by the most abundant cells in a sample, leading to the underrepresentation of rare cell types.Recent developments in single-cell epigenetic profiling methods have mitigated these issues, although large parts of the epigenome, such as histone modifications and chromatin interactions, remain relatively under-explored at the single-cell level. In my thesis, I developed a set of sequencing tools to link transcriptional outcomes with epigenetic information, including histone modifications and chromatin interactions, in single cells. This enables understanding the cell type-specific roles of under-explored repressive elements and chromatin structures in both healthy and diseased mammalian brains. In Chapter 1, I describe Droplet Paired-Tag, a droplet-based single-cell joint profiling method for histone modifications and transcriptomes. This method addresses the challenges of current single-cell histone modification mapping techniques by offering shorter handling times, a simplified protocol, and higher data quality. Through this approach, I successfully annotated the chromatin states of known cis-regulatory elements (cCREs) and predicted both active and repressive cCREs in various cell types in the adult mouse cerebral cortex. In Chapter 2, I detail the development of Droplet Hi-C, a high-throughput method for profiling chromatin conformation in single cells. This technique not only enables large-scale analysis of cell type-specific chromatin structures in heterogeneous tissues, such as the mouse brain, but also has been pivotal in detecting copy number and structural variations, including extrachromosomal DNA (ecDNA), in brain cancer cells and glioblastoma. Furthermore, I demonstrate that Droplet Hi-C can be extended to a multiomics approach called Paired Hi-C, which facilitates the simultaneous analysis of gene expression and chromatin conformation in the same cell. This advancement provides an opportunity to connect ecDNA copy number with gene expression variations in brain cancer cells under drug treatment. In summary, my thesis demonstrates a strong commitment to advancing our understanding of the epigenome through the development of cutting-edge single-cell technologies. By applying these methods to human brain samples, I aim to decode the complexities of gene regulation and chromatin dynamics underlying various developmental and disease trajectories.