Project Summary Understanding how genes' activities are controlled is crucial for elucidating the basic operating rules of biology and molecular mechanisms of diseases. Recent innovations in single-cell genomic technologies have opened the door to analyzing a variety of functional genomic features in individual cells. These technologies enable scientists to systematically discover unknown cell subpopulations in complex tissue and disease samples, and allow them to reconstruct a sample's gene regulatory landscape at an unprecedented cellular resolution. Despite these promising developments, many challenges still exist and must be overcome before one can fully decode gene regulation at the single-cell resolution. In particular, current technologies lack the ability to accurately measure the activity of each individual cis-regulatory element (CRE) in a single cell. They also cannot measure all functional genomic data types in the same cell. Moreover, the prevalent technical biases and noises in single-cell genomic data make computational analysis non-trivial. With rapid growth of data, lack of computational tools for data analysis has become a rate-limiting factor for effective applications of single-cell genomic technologies. The objective of this proposal is to develop computational and statistical methods and software tools for mapping and analyzing gene regulatory landscape using single-cell genomic data. Our Aim 1 addresses the challenge of accurately measuring CRE activities in single cells using single-cell regulome data. Regulome, defined as the activities of all cis-regulatory elements in a genome, contains crucial information for understanding gene regulation. The state-of-the-art technologies for mapping regulome in a single cell produce sparse data that cannot accurately measure activities of individual CREs. We will develop a new computational framework to allow more accurate analysis of individual CREs' activities in single cells using sparse data. Our Aim 2 addresses the challenge of collecting multiple functional genomic data types in the same cell. We will develop a method that uses single-cell RNA sequencing (scRNA-seq), the most widely used single-cell functional genomic technology, to predict cells' regulatory landscape. Since most scRNA-seq datasets do not have accompanying single-cell data for other -omics data types, our method will also significantly expand the utility and increase the value of scRNA- seq experiments. Our Aim 3 addresses the challenge of integrating different data types generated by different single-cell genomic technologies from different cells. We will develop a method to align single-cell RNA-seq and single-cell regulome data to generate an integrated map of transcriptome and regulome. Upon completion of this proposal, we will deliver our methods through open-source software tools. These tools will be widely useful for analyzing and integrating single-cell regulome and transcriptome data. By addressing sev...