PROJECT SUMMARY This project proposes to develop and refine computational tools to advance the safety and efficacy of genome editing, addressing the critical need for precise quantification and prediction of on- and off-target effects. Genome editing technologies have the potential to revolutionize the treatment and prevention of disease by enabling precise modifications to DNA sequences. However, their clinical application is currently limited by significant challenges in detecting and avoiding off-target effects, as well as quantifying and understanding the heterogeneity of on-target edits. In Aim 1, we will enhance the capabilities of our recently developed tool, CRISPRme, which predicts off-target effects across diverse human genomes. We propose to develop a new tool, CRISPRus, which will leverage variation graphs, complete genomes, and pangenomes to provide a more comprehensive understanding of potential off-target effects. Aim 2 will focus on the characterization of complex on-target editing outcomes. We will develop and refine tools to enable accurate quantification from long-read and single-anchor sequencing methods. This will provide a more detailed and nuanced understanding of the outcomes of genome editing, including the potential for complex, multi-site edits. In Aim 3, we will tackle the statistical challenges inherent in quantifying genome editing events. We propose to develop DE-CRISPR, a comprehensive framework that will provide simulations and statistical models to assess the significance of editing events at the locus, repair-pathway, and individual allele levels. This will enable researchers to design experiments with adequate power to detect editing events and to interpret their results with greater confidence and precision. Accompanying these aims, we will produce comprehensive benchmark datasets, including of 6 editing approaches in 6 donors; comparing nuclease, base, and prime editing; comparing approaches across a range of anticipated specificity; targeting primary clinically relevant cells of unknown genotype (a real-world scenario) and cell lines with complete telomere-to-telomere genome assemblies described. The deliverables of this project, a comprehensive toolkit of computational tools including CRISPRus, CRISPRessoPE, CRISPRuni, CRISPRlungo and DE-CRISPR, will empower researchers to more safely and effectively harness the power of genome editing technologies. This work is essential to ensuring the robustness and reproducibility of genome editing, ultimately promoting the safety and efficacy of therapeutic development. By addressing current limitations in genome editing off-target prediction, complex edit characterization, and statistical analysis, we aim to keep pace with advances in genome editing technology and contribute to the realization of its full therapeutic potential.