Large-Scale In Vivo Functional Characterization of the Human Cistrome



 DESCRIPTION (provided by applicant): The in vitro functions of thousands of cis-elements have been characterized using massively parallel reporter assays. However, the in vivo characterization of cis-element functions has been limited to a handful of mammalian enhancers using mouse models. Recently, genome-wide CRISPR/Cas9 knockout screening has proven to be an efficient approach for assessing the function of genomic DNAs and presenting an opportunity to study in vivo cis-element functions on a large scale. We have developed a statistical model to predict gRNA efficiency for the optimization of CRISPR/Cas9 screens, as well as a computational pipeline for the analysis of the screen data. In addition we have performed pilot genome- wide CRISPR/Cas9 knockout screens in several cell lines and have identified known genes essential cell growth. We have demonstrated expertise in studying gene transcriptional regulation through genome-wide analysis of protein-DNA interactions and chromatin accessibility. In this proposal, we hypothesize that the simultaneous lentiviral delivery of two guide RNAs (gRNA) flanking a cis-regulatory element can efficiently knockout (KO) the element, and that CRISPR/Cas9 KO screens are an efficient approach for the large-scale in vivo functional characterization of human cis-elements. We propose to develop the experimental and computational approaches for high-throughput cistrome CRISPR/Cas9 deletion screens to elucidate the regulatory mechanisms of mammalian cis-element in vivo functions and expand our knowledge on transcriptional regulation in normal physiology and diseases. Specifically we propose to 1) use CRISPR/Cas9 knockout screens to identify transcription factors and chromatin regulators in eight human cell lines that have strong effect on cell growth; 2) conduct CRISPR/Cas9 knockout screens on putative cis-regulatory elements to identify elements with strong effects on gene expression and cell growth or survival; 3) computationally model in vivo cistrome function, experimentally validate the model and create a Cistrome annotation web server. At the conclusion of these studies, we will have developed the experimental techniques and computational tools for continued investigation of in vivo cis-element functions, and expanded our knowledge on the mechanism of cell-specific gene transcriptional regulation. The resulting resource will improve interpretation of the function of disease associated somatic mutations or germline variants in the non-coding regions.