Victor Zhivich

Genome-scale CRISPR-Cas9 viability screens performed in cancer cell lines provide a systematic approach to identify cancer dependencies and new therapeutic targets. As multiple large-scale screens become available, a formal assessment of the reproducibility of these experiments becomes necessary. We analyze data from recently published pan-cancer CRISPR-Cas9 screens performed at the Broad and Sanger Institutes. Despite significant differences in experimental protocols and reagents, we find that the screen results are highly concordant across multiple metrics with both common and specific dependencies jointly identified across the two studies. Furthermore, robust biomarkers of gene dependency found in one data set are recovered in the other. Through further analysis and replication experiments at each institute, we show that batch effects are driven principally by two key experimental parameters: the reagent library and the assay length. These results indicate that the Broad and Sanger CRISPR-Cas9 viability screens yield robust and reproducible findings.

The CRISPR-Cas9 system has revolutionized gene editing both on single genes and in multiplexed loss-of-function screens, enabling precise genome-scale identification of genes essential to proliferation and survival of cancer cells. However, previous studies reported that an anti-proliferative effect of Cas9-mediated DNA cleavage confounds such measurement of genetic dependency, particularly in the setting of copy number gain 1-4 . We performed genome-scale CRISPR-Cas9 essentiality screens on 342 cancer cell lines and found that this effect is common to all lines, leading to false positive results when targeting genes in copy number amplified regions. We developed CERES, a computational method to estimate gene dependency levels from CRISPR-Cas9 essentiality screens while accounting for the copy-number-specific effect, as well as variable sgRNA activity. We applied CERES to sets of screens performed with different sgRNA libraries and found that it reduces false positive results and provides meaningful estimates of sgRNA activity. As a result, the application of CERES improves confidence in the interpretation of genetic dependency data from CRISPR-Cas9 essentiality screens of cancer cell lines.

Abstract Genome-scale CRISPR-Cas9 viability screens performed in cancer cell lines provide a systematic approach to identify cancer dependencies and new therapeutic targets. As multiple large-scale screens become available, a formal assessment of the reproducibility of these experiments becomes necessary. We analyzed data from recently published pan-cancer CRISPR-Cas9 screens performed at the Broad and Sanger institutes. Despite significant differences in experimental protocols and reagents, we found that the screen results are highly concordant across multiple metrics with both common and specific dependencies jointly identified across the two studies. Furthermore, robust biomarkers of gene dependency found in one dataset are recovered in the other. Through further analysis and replication experiments at each institute, we found that batch effects are driven principally by two key experimental parameters: the reagent library and the assay length. These results indicate that the Broad and Sanger CRISPR-Cas9 viability screens yield robust and reproducible findings.