Evaluation and Design of Genome-Wide CRISPR/SpCas9 Knockout Screens.
Hart T, Tong AHY, Chan K, Van Leeuwen J, Seetharaman A, Aregger M, Chandrashekhar M, Hustedt N, Seth S, Noonan A, Habsid A, Sizova O, Nedyalkova L, Climie R, Tworzyanski L, Lawson K, Sartori MA, Alibeh S, Tieu D, Masud S, Mero P, Weiss A, Brown KR, Usaj M, Billmann M, Rahman M, Constanzo M, Myers CL, Andrews BJ, Boone C, Durocher D, Moffat J. · 2017
Read Evaluation and Design of Genome-Wide CRISPR/SpCas9 Knockout Screens. by Hart T, Tong AHY, Chan K, Van Leeuwen J, Seetharaman A, Aregger M, Chandrashekhar M, Hustedt N, Seth S, Noonan A, Habsid A, Sizova O, Nedyalkova L, Climie R, Tworzyanski L, Lawson K, Sartori MA, Alibeh S, Tieu D, Masud S, Mero P, Weiss A, Brown KR, Usaj M, Billmann M, Rahman M, Constanzo M, Myers CL, Andrews BJ, Boone C, Durocher D, Moffat J. (2017) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5555476.txt.
Introduction
Hart and co-authors present a methods study on CRISPR knockout screens. Such screens switch off genes across the whole genome in human cells. They show which genes cells need to stay alive. Results vary a lot with library design and protocol, though.
The team reanalyzed 17 genome-scale screens from three research groups. Their work expanded the set of human core essential genes from 360 to 684. They also studied what makes a guide RNA work well. From those lessons they built an optimized library, Toronto KnockOut version 3. That library performed strongly in fresh screens. Readers get both a reference gene set and a better tool for cell studies.
Catalog record
| CREATOR | Hart T, Tong AHY, Chan K, Van Leeuwen J, Seetharaman A, Aregger M, Chandrashekhar M, Hustedt N, Seth S, Noonan A, Habsid A, Sizova O, Nedyalkova L, Climie R, Tworzyanski L, Lawson K, Sartori MA, Alibeh S, Tieu D, Masud S, Mero P, Weiss A, Brown KR, Usaj M, Billmann M, Rahman M, Constanzo M, Myers CL, Andrews BJ, Boone C, Durocher D, Moffat J. |
|---|---|
| YEAR | 2017 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Evaluation and Design of Genome-Wide CRISPR/SpCas9 Knockout Screens By Traver Hart, Amy Hin Yan Tong, Katie Chan, Jolanda Van Leeuwen, Ashwin Seetharaman, Michael Aregger, Megha Chandrashekhar, Nicole Hustedt, Sahil Seth, Avery Noonan, Andrea Habsid, Olga Sizova Published in: G3: Genes|Genomes|Genetics DOI: 10.1534/g3.117.041277 ABSTRACT The adaptation of CRISPR/SpCas9 technology to mammalian cell lines is transforming the study of human functional genomics. Pooled libraries of CRISPR guide RNAs (gRNAs) targeting human protein-coding genes and encoded in viral vectors have been used to systematically create gene knockouts in a variety of human cancer and immortalized cell lines, in an effort to identify whether these knockouts cause cellular fitness defects. Previous work has shown that CRISPR screens are more sensitive and specific than pooled-library shRNA screens in similar assays, but currently there exists significant variability across CRISPR library designs and experimental protocols. In this study, we reanalyze 17 genome-scale knockout screens in human cell lines from three research groups, using three different genome-scale gRNA libraries. Using the Bayesian Analysis of Gene Essentiality algorithm to identify essential genes, we refine and expand our previously defined set of human core essential genes from 360 to 684 genes. We use this expanded set of reference core essential genes, CEG2, plus empirical data from six CRISPR knockout screens to guide the design of a sequence-optimized gRNA library, the Toronto KnockOut version 3.0 (TKOv3) library. We then demonstrate the high effectiveness of the library relative to reference sets of essential and nonessential genes, as well as other screens using similar approaches. The optimized TKOv3 library, combined with the CEG2 reference set, provide an efficient, highly optimized platform for performing and assessing gene knockout screens in human cell lines. The generation of gene knockouts is a cornerstone of functional genomics research. The application of CRISPR technology to induce targeted DNA double-strand breaks in mammalian cells (Jinek et al. 2012), coupled with the ability of the endogenous cellular DNA repair machinery to introduce indels when repairing these lesions, has led to the rapid development of pooled-library CRISPR knockout screens in mammalian cells for functional genomics, chemogenomics, the identification of cancer cell vulnerabilities, and other applications (Hart et al. 2015; Koike-Yusa et al. 2014; Shalem et al. 2014; Wang et al. 2014, 2015, 2017; Parnas et al. 2015; Tzelepis et al. 2016; Aguirre et al. 2016). CRISPR screens represent a major advance over pooled-library shRNA screens (Evers et al. 2016), the prior state-of-the-art in mammalian functional screening approaches, in both sensitivity and specificity. The current designs of large-scale CRISPR experiments benefited from the many lessons learned in shRNA screening. In particular, the design of early CRISPR libraries to include several guide RNAs (gRNAs) targeting each gene has been driven by experience with pooled-library shRNA screens (Kaelin, 2012; Echeverri et al. 2006), as well as the unknowns surrounding the application of CRISPR technology in human cells on a large scale. Integrated analysis of multiple reagents targeting the same gene should overcome the noise introduced by variable reagent effectiveness and the unknown frequency and impact of off-target effects. With several panels of whole-genome cell-line screens published (Aguirre et al. 2016; Hart et al. 2015; Tzelepis et al. 2016; Wang et al. 2015, 2017), the opportunity now exists to undertake a meta-analysis as a means to uncover the drivers of screen quality and variability. Thus, we reanalyzed sets of CRISPR screens conducted in adherent and suspension cell lines, using three different large-scale libraries, and evaluated each for quality and consistency. Based on these observations, we refined our list of core essential genes (CEG), i.e., the set of genes that are likely to be essential in all human cell lines. We evaluated the impact of experimental design, including library size, number of replicates, and use of nontargeting controls, on screen performance. Finally, we derived a sequence signature for highly effective gRNAs and designed an optimized, genome-scale CRISPR library for efficient screening of human cell lines. MATERIALS AND METHODS The supplemental file HART_data_and_python_notebooks.tgz is available at http://tko.ccbr.utoronto.ca/ for download, and contains python notebooks and all required data to generate the figures presented here. As such, it contains a near-complete, granular description of the computational methods applied in this study. Detailed experimental methods and a summary of computational methods are included below. Analysis of screens from various libraries gRNA readcount data were downloaded from Hart et al. (2015), Koike-Yusa et al. (2014), Tzelepis et al. (2016), and Wang et al. (2015). Fold-changes were calculated by normalizing each sample to 10 million reads and calculating log2 (experimental/control) for each replicate of each sample at each timepoint. Control was either the gRNA counts from the genomic DNA collected after infection (TO) or library plasmid pool, depending on the experimental design. A pseudocount of 0.5 reads was added to all readcounts to prevent discontinuities from zeros. gRNA with <30 reads at the T0 timepoint were excluded from the fold-change calculation. Fold-changes were processed with the Bayesian Analysis of Gene Essentiality (BAGEL) algorithm (Hart and Moffat 2016), using the essential and nonessential training sets defined in Hart et al. (2014). The resulting Bayes Factors (BFs) for all screens are included in Supplemental Material, Table S1. After the Core Essential Genes 2.0 (CEG2) set was defined, BFs for all screens were recalculated using this new reference set (Table S2). Identification of CEG2 Of the 17 gRNA screens initially evaluated, three were withheld for later evaluation and analysis. Two others were excluded for relatively poor performance. For the remaining 12 screens, the BF and the number of gRNAs targeting the gene were considered. Note that the number of gRNAs may vary by cell line and by library since only gRNAs with >30 reads in the T0 control sample were used for each cell line screen. A gene was defined as effectively assayed if it was targeted by at least three gRNAs in a given screen. The CEG2 set was defined as genes effectively assayed in at least seven cell lines, with a strict BF threshold of ≥6 in 85% of cell lines in which they were assayed. Since most genes were assayed in either seven or 12 cell lines, this effectively means that core essentials are hits in six of seven or 11 of 12 screens. The strict BF ≥ 6 threshold corresponds to a posterior probability of ∼90%. To calculate posterior odds from a BF, it is necessary to multiply by a ratio of priors. Based on significant empirical screening evidence, we estimate ∼10% of genes to be essential in a given cell line; the prior ratio P(essential)/P(nonessential) is therefore 0.1/0.9, which in log2 is ∼−3. Therefore, a BF of three corresponds to a posterior log odds of ∼0, or posterior probability of essentiality of 50%, the threshold we generally apply for identifying essential genes (provided a FDR threshold is also met). A BF of six therefore has posterior log odds of three, or posterior probability of ∼90%. Evaluation of gRNA per gene The Sabatini library in Wang et al. (2015) contains 10 gRNA per gene. For each of the five screens in four cell lines (KBM7 is screened twice), a subset of guides was randomly selected and BFs for all genes were calculated from the subset. This process was repeated 10 times