CRISPR-Cas9 viability screens are increasingly performed at a genome-wide scale across large panels of cell lines to identify new therapeutic targets for precision cancer therapy. Integrating the datasets resulting from these studies is necessary to adequately represent the heterogeneity of human cancers and to assemble a comprehensive map of cancer genetic vulnerabilities. Here, we integrated the two largest public independent CRISPR-Cas9 screens performed to date (at the Broad and Sanger institutes) by assessing, comparing, and selecting methods for correcting biases due to heterogeneous single-guide RNA efficiency, gene-independent responses to CRISPR-Cas9 targeting originated from copy number alterations, and experimental batch effects. Our integrated datasets recapitulate findings from the individual datasets, provide greater statistical power to cancer- and subtype-specific analyses, unveil additional biomarkers of gene dependency, and improve the detection of common essential genes. We provide the largest integrated resources of CRISPR-Cas9 screens to date and the basis for harmonizing existing and future functional genetics datasets.

CRISPR loss of function screens are powerful tools to interrogate biology but exhibit a number of biases and artifacts that can confound the results. Here, we introduce Chronos, an algorithm for inferring gene knockout fitness effects based on an explicit model of cell proliferation dynamics after CRISPR gene knockout. We test Chronos on two pan-cancer CRISPR datasets and one longitudinal CRISPR screen. Chronos generally outperforms competitors in separation of controls and strength of biomarker associations, particularly when longitudinal data is available. Additionally, Chronos exhibits the lowest copy number and screen quality bias of evaluated methods. Chronos is available at https://github.com/broadinstitute/chronos.

CRISPR-Cas9 viability screens are being increasingly performed at a genome-wide scale across large panels of cell lines to identify new therapeutic targets for precision cancer therapy. Integrating the datasets resulting from these studies is necessary to adequately represent the heterogeneity of human cancers, and to assemble a comprehensive map of cancer genetic vulnerabilities that might be exploited therapeutically. Here, we integrated the two largest independent CRISPR-Cas9 screens performed to date (at Broad and Sanger institutes), by assessing and selecting methods for correcting technology-specific biases and batch effects arising from differences in the underlying experimental protocols. Our integrated datasets recapitulate findings from the individual ones, provide larger statistical power allowing novel cancer-and subtype-specific analyses, unveil additional biomarkers of gene dependency, and improve the detection of common essential genes. Finally, we provide the largest integrated resources of CRISPR-Cas9 screens to date and the basis for harmonizing existing and future functional genetics datasets and assembling large cross-study cancer dependency maps.Cancer is a complex disease that can arise from multiple different genetic alterations.The alternative mechanisms by which cancer can evolve result in a large amount of heterogeneity between patients, with the vast majority of them still not benefiting from approved targeted therapies 1 . In order to identify and prioritize new potential therapeutic targets for precision cancer therapy, analyses of cancer vulnerabilities at a genome-wide scale and across large panels of in vitro cancer models are being increasingly performed 2-11 . This has been facilitated by recent advances in genome editing technologies allowing unprecedented precision and scale via CRISPR-Cas9 screens. To date, two large pan-cancer CRISPR-Cas9 screens have been independently performed by the Broad and Sanger institutes 2,12 . The two institutes have also joined forces in a collaborative endeavor with the aim of assembling a joint comprehensive map of all the intracellular genetic dependencies and vulnerabilities of cancer: the Cancer Dependency Map (DepMap) 13,14 .Despite the two generated datasets containing so far data from over 300 cell lines each, detecting all cancer dependencies has been estimated to require the analysis of thousands of cancer models 3 . Consequently, the integration of these two datasets will be key for the DepMap and other projects aiming at systematically probing cancer dependencies.This will provide a more comprehensive representation of heterogeneous cancer types as well as an increased sample size to support the use of statistical/machine-learning methods to identify molecular features associated with differential gene dependencies. This will form the basis for the development of effective new therapies with associated biomarkers for patient stratification 15 . Furthermore, designing robust standards and computational protocols for the integration of t...

Reducing disparities is critical to promote equity of access to precision treatments for all patients with cancer. While socioenvironmental factors are a major driver behind such disparities, biological differences also are likely to contribute. The prioritization of cancer drug targets is foundational for drug discovery, yet whether ancestry-related signals in target discovery pipelines exist has not been systematically explored due to the absence of data at the appropriate scale. Here, we analyzed data from 611 genome-scale CRISPR/Cas9 viability experiments in human cell line models as part of the Cancer Dependency Map to identify ancestry-associated genetic dependencies. Surprisingly, we found that most putative associations between ancestry and dependency arose from artifacts related to germline variants that are present at different frequencies across ancestry groups. In 2-5% of genes profiled in each cellular model, germline variants in sgRNA targeting sequences likely reduced cutting by the CRISPR/Cas9 nuclease. Unfortunately, this bias disproportionately affected cell models derived from individuals of recent African descent because their genomes tended to diverge more from the consensus genome typically used for CRISPR/Cas9 guide design. To help the scientific community begin to resolve this source of bias, we report three complementary methods for ancestry-agnostic CRISPR experiments. This report adds to a growing body of literature describing ways in which ancestry bias impacts cancer research in underappreciated ways.

Socioeconomic factors and discrimination play major roles in variable cancer treatment outcomes across individuals from different ancestral backgrounds. Tumors from patients of different ancestry groups also have divergent patterns of somatic and germline alterations. This led us to hypothesize that these molecular differences may also contribute to the variable treatment outcomes observed in the clinic. We hypothesized that the ancestry of cell lines would impact their genetic dependencies. To test this hypothesis, we leveraged The Cancer Dependency Map (DepMap) which has performed genome-wide CRISPR screens across >1,000 cell lines and >30 cancer types. We first leveraged variant calls from WES/WGS to infer cell line ancestry, then correlated these ancestries with DepMap gene dependency scores. This analysis was underpowered to detect differences in cell lines of African, American, and South Asian descent, since cell lines from these ancestry groups are poorly represented in DepMap, and in cancer research models in general. We were, however, able to detect 71 gene dependencies that were associated with either European or East Asian ancestry. Since different ancestry groups have divergent patterns of germline alterations, we reasoned that specific germline alterations may result in ancestry-associated dependencies. Surprisingly, we identified cis-QTLs for >75% of the ancestry-associated genes. We originally hypothesized that these variants would alter the function of the encoded protein, but instead found that these variants mapped to the targeting sequences of the sgRNA. This suggests that ancestry-associated mismatches within sgRNA targeting sequences can preclude Cas9-mediated genome editing. To understand this problem systematically, we mapped the germline variants that were catalogued in gnomAD to multiple genome-wide CRISPR libraries. The fraction of affected genes differed from library to library and was typically between 2-5%, but we found that individuals of African descent were consistently more affected by this problem across all CRISPR libraries. This is important because it suggests that cell lines of African descent have a higher rate of false negatives in all CRISPR-based experiments. In total, we identified a subset of genes that have ancestry-associated dependency profiles. Most of these genes are the result of ancestry-associated mismatches within the sgRNA targeting sequences. However, many of these genes are not, and require further investigation to understand the influence of ancestry on these genetic dependencies. Citation Format: Sean Alexander Misek, Aaron Fultineer, Jeremie Kalfon, Javad Noorbakhsh, Isabella Boyle, Joshua Dempster, Lia Petronio, Katherine Huang, James McFarland, Rameen Beroukhim, Jesse Boehm. Ancestry bias in CRISPR guide design impedes discovery of genetic dependencies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2173.