Massively Parallel CRISPR‐Based Genetic Perturbation Screening at Single‐Cell Resolution

Abstract

The clustered regularly interspaced short palindromic repeats (CRISPR)‐based genetic screening has been demonstrated as a powerful approach for unbiased functional genomics research. Single‐cell CRISPR screening (scCRISPR) techniques, which result from the combination of single‐cell toolkits and CRISPR screening, allow dissecting regulatory networks in complex biological systems at unprecedented resolution. These methods allow cells to be perturbed en masse using a pooled CRISPR library, followed by high‐content phenotyping. This is technically accomplished by annotating cells with sgRNA‐specific barcodes or directly detectable sgRNAs. According to the integration of distinct single‐cell technologies, these methods principally fall into four categories: scCRISPR with RNA‐seq, scCRISPR with ATAC‐seq, scCRISPR with proteome probing, and imaging‐based scCRISPR. scCRISPR has deciphered genotype–phenotype relationships, genetic regulations, tumor biological issues, and neuropathological mechanisms. This review provides insight into the technical breakthrough of scCRISPR by systematically summarizing the advancements of various scCRISPR methodologies and analyzing their merits and limitations. In addition, an application‐oriented approach guide is offered to meet researchers’ individualized demands.

PerturBase: a comprehensive database for single-cell perturbation data analysis and visualization

“…It should be noted that if no perturbation meets these criteria, the dataset is not subjected to further analysis; (4) PerturBase adopts the global scaling normalization method in Scanpy to scale the expression in each cell to 10,000, followed by logarithmic transformation; (5) after normalization, PerturBase adopts “highly variable genes” with default parameter in Scanpy to identify highly variable features for scPerturbation data. To strike a balance between computational efficiency and data retention, PerturBase maintains a gene count ranging from a minimum of 2000 to a maximum of 4000 genes; (6) PerturBase performs principal component analysis (PCA) on the top variable genes, adhering to default parameters (n_components = 50) to reduce dataset dimensionality; (7) After dimension reduction, PerturBase performs clustering based on the leiden algorithm with default parameters; (8) For user convenience, we employed clusterProfiler to obtain gene symbol, Entrez and Ensembl IDs for genes in each dataset, making them readily accessible to users within their dataset of interest(31).…”

Toward subtask decomposition-based learning and benchmarking for genetic perturbation outcome prediction and beyond

“…The transcriptional outcomes of a single-cell genetic perturbation reveal fundamental insights into cell functions and how a cell responds to external interventions. Various high-throughput sequencing technologies, including single-cell RNA sequencing 1 (scRNA-seq) and single-cell Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) genetic screening methods 2 (for example, Perturb-seq 3 ), have enabled the dissection of the transcriptional responses of individual cells to specific genetic perturbations, facilitating the uncovering of the roles and interactions of genes 4 . This has profound implications for investigating the regulatory mechanisms of genes and unraveling effective therapeutics in biomedical studies [5][6][7] .…”

A mini-review on perturbation modelling across single-cell omic modalities

“…On the other hand, there is also a burgeoning number of single-cell perturbational screens , which we will briefly address in this section. In this section, we concisely introduce the fundamental single-cell technologies underpinning this field, and we encourage the readers to explore an excellent review by Cheng et al [6] , which delves more in-depth into this particular subject.…”