Dataset Information

Improved prediction of bacterial CRISPRi guide efficiency from depletion screens through mixed-effect machine learning and data integration

ABSTRACT: CRISPR interference (CRISPRi), the targeting of a catalytically dead Cas protein to block transcription, is the leading technique to silence gene expression in bacteria. However, design rules for CRISPRi remain poorly defined, limiting predictable design for gene interrogation, pathway manipulation, and high-throughput screens. Here we develop a best-in-class prediction algorithm for guide silencing efficiency by systematically investigating factors influencing guide depletion in multiple genome-wide essentiality screens, with the surprising discovery that gene-specific features such as transcriptional activity substantially impact prediction of guide activity. Accounting for these features as part of algorithm development allowed us to develop a mixed-effect random forest regression model that provides better estimates of guide efficiency than existing methods, as demonstrated in an independent saturating screen. We further applied methods from explainable AI to extract interpretable design rules from the model, such as sequence preferences in the vicinity of the PAM distinct from those previously described for genome engineering applications. Our approach provides a blueprint for the development of predictive models for CRISPR technologies where only indirect measurements of guide activity are available.

ORGANISM(S): Escherichia coli

PROVIDER: GSE196911 | GEO | 2023/12/19

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Background: The CRISPR/Cas9 toolbox has recently been expanded to include approaches for modulating gene expression. To successfully build on this work, and apply it for answering biological questions, it is important to establish it in a broad range of circumstances. Genome-scale CRISPR interference (CRISPRi) has been used in human cells lines, however the rules for designing effective guide RNAs (gRNAs) in different organisms are not well known. We sought to determine rules that determine gRNA effectiveness at transcriptional repression in Saccharomyces cerevisiae. Results: We created an inducible single plasmid CRISPRi system for gene repression in yeast, and used it to analyze fitness effects of gRNAs under 18 small molecule treatments. Our approach correctly identified previously-described chemical-genetic interactions, as well as a new mechanism of suppressing fluconazole toxicity by repression of the ERG25 gene. Assessment of multiple target loci across treatments allowed us to determine generalizable features associated with gRNA efficacy. Guides that target regions with low nucleosome occupancy and high chromatin accessibility were clearly more effective. We also found the best region to target gRNAs was between the transcription start site (TSS) and 200bp upstream of the TSS. Finally, unlike nuclease-proficient Cas9 in human cells, point mutations were tolerated equally well by truncated (18 nt specificity sequence) and full length (20 nt) gRNAs, however, 18 nt gRNAs were generally less potent than full length gRNAs. Conclusions: Our results establish a powerful functional genomics screening method, provide rules for designing effective gRNAs for gene repression, and show that 18 nt and 20 nt gRNAs exhibit similar tolerance to mismatches in the target sequence. These findings will enable effective library design and genome-wide screening in many genetic backgrounds. An expression construct was created for inducible CRISPRi in yeast. Key features include ORFs expressing dCas9-Mxi1 and the tetracycline repressor (TetR), as well as a tetracycline inducible gRNA locus containing the RPR1 promoter with a TetO site, a NotI site for cloning new gRNA specificity sequences, and the constant part of the gRNA. When yeast containing this plasmid are grown in the absence of anhydrotetracycline (ATc) TetR binds the gRNA promoter and prevents PolIII from binding and transcribing the gRNA. This in turn prevents dCas9-Mxi1 from binding the target site. In the presence of ATc, TetR dissociates and gRNA is expressed, allowing dCas9-Mxi1 to bind its target locus, and repress gene expression. gRNA libraries were cloned into this construct and transformed into yeast to create pools. Experiments were conducted in which yeast pools were grown in inducing (+ATc) and non-inducing conditions (-ATc) in the presence of different drugs. After multiple generations of growth in these conditions, yeast plasmids were minipreped and the gRNA locus was PCRed and sequenced via MiSeq. Counts of each gRNA were compared in different conditions.

Dataset Information

Improved prediction of bacterial CRISPRi guide efficiency from depletion screens through mixed-effect machine learning and data integration

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure