Computational Technology for Single-Cell Functional Genomics - Abstract Genome-wide association studies have unveiled millions of single nucleotide polymorphisms (SNPs) associated with common diseases. Yet, the functional mechanism of most SNPs in these diseases still needs to be discovered. My recent studies suggested that hundreds of thousands of SNPs strongly associate with ligand gene expression in the GTEx dataset, suggesting the impacts of these SNPs on many cell types in a tissue niche via ligand-receptor interactions. However, the lack of robust technology that adequately considers the context of tissue niches constrains the systematic characterization of SNP functions in health and diseases. This project, empowered by functional genomics approaches, aims to address this challenge. I will develop robust computational technologies for single-cell functional genomics, allowing comprehensive identification of SNPs regulating tissue niches and elucidating their regulatory roles in health and diseases. In Aim 1 (K99 phase), I will develop computational frameworks with deep learning models to decode the epigenetics response of a cell to intercellular communications with other cells. The performance of my algorithm will be systematically validated by RNA-seq and ATAC-seq profiles under control and ligand treatment conditions. In Aim 2 (K99 and R00 phase), I will develop robust statistical methods to identify the impact of SNPs on the tissue niches via ligand-receptor communications. The effects of SNPs on ligand expression will be elucidated by integrating the SNPs with genome-wide transcription factor binding sites and chromatin status. In Aim 3 (R00 phase), I will employ robust computational technologies to identify clinically relevant SNPs that regulate tissue niches via intercellular ligand-receptor communications. I will also validate the SNP functions by prime editing, focusing on endothelial cells and adipocytes in adipose tissue, which are tightly associated with many diseases, including obesity, diabetes and complications, and cardiovascular diseases. This study will lay a solid foundation for future research as an independent scientist focusing on genetic diseases. My extensive experience in epigenetics, intercellular communication modeling, and single- cell data analysis positions me uniquely to execute this proposal. The project will be supervised by an interdisciplinary team, including Dr. Kaifu Chen for bioinformatics, Dr. Yu-Hua Tseng for metabolic diseases, and five collaborators with expertise in deep learning, big data integration, statistical genetics, and prime editing. The training will also include professional skills such as scientific writing, communication, and mentoring. The outstanding research environment at Boston Children’s Hospital and Harvard Medical School will further enhance the execution of the proposed study. Completing this training and proposal during the K99 phase will well prepare me for a future role as an independent researcher in translational genetics and genomics research, ensuring a seamless transition into the R00 phase.
