Abstract:Recent single-cell CRISPR screening experiments have combined the advances of genetic editing and single-cell technologies, leading to transcriptome-scale readouts of responses to perturbations at single-cell resolution. An outstanding question is how to efficiently identify heterogeneous causal effects of perturbations using these technologies. Here we present scCAPE, a tool designed to facilitate causal analysis of heterogeneous perturbation effects at the single-cell level. scCAPE disentangles perturbation effects from the inherent cell-state variations and provides nonparametric inferences of perturbation effects at single-cell resolution, permitting a range of downstream tasks including perturbation effect analysis, genetic interaction analysis, perturbation clustering and prioritizing. We benchmarked scCAPE through simulation studies and real datasets to evaluate its performance in characterizing latent confounding factors and accuracy in estimating heterogeneous perturbation effects. The application of scCAPE identified novel heterogeneous genetic interactions among erythroid differentiation drivers. For example, our analysis pinpointed the role of the synergistic interaction between CBL and CNN1 in the S phase.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: how to efficiently identify heterogeneous causal effects in single - cell gene - editing perturbation experiments. Specifically, the research aims to develop a tool that can analyze the heterogeneous impacts of gene perturbations at single - cell resolution and separate perturbation effects from inherent cell - state changes, thereby providing support for downstream tasks such as gene - interaction analysis, perturbation clustering, and prioritization. ### Problem Background In recent years, single - cell CRISPR screening technology, which combines the advantages of gene editing and single - cell sequencing, enables researchers to obtain transcriptome - scale perturbation - response readings at the single - cell level. However, how to effectively identify heterogeneous causal effects in these data remains a challenge. Traditional pseudo - bulk methods assume that all cells respond homogeneously to perturbations, which ignores the potential heterogeneous landscape and limits in - depth understanding of regulatory mechanisms. ### Research Objectives To address this challenge, the authors propose scCAPE (Single - Cell Causal Analysis of Perturbation Effects), a framework specifically designed for causal analysis at the single - cell level. The main objectives of scCAPE include: 1. **Analyze Heterogeneous Perturbation Effects**: Estimate perturbation effects at single - cell resolution through non - parametric inference. 2. **Separate Perturbation Effects from Cell - State Variation**: Isolate causal changes caused by perturbations from the transcriptome profile, excluding the influence of other factors. 3. **Provide Rich Downstream Applications**: Including gene - interaction analysis, perturbation clustering, and prioritization, etc. ### Solutions scCAPE achieves its objectives through the following steps: 1. **Autoencoder and Adversarial Training**: Use an autoencoder with adversarial training to decouple perturbation effects and inherent cell - state changes, generating a "ground - state space" that does not contain perturbation information but retains the cell's intrinsic state. 2. **Causal Forest Estimation**: Based on the ground - state space, use the causal forest to iteratively split cells, maximizing the heterogeneity of perturbation effects, thereby estimating the perturbation effect of each cell and its asymptotic variance. 3. **Statistical Testing and Functional Annotation**: Conduct statistical tests on the estimated perturbation effects to determine their significance, and perform functional annotation through biomarker association and gene - set enrichment analysis. ### Application Examples The application of scCAPE on simulated data and real - world datasets demonstrates its powerful performance. For example, in datasets of human CD8+ T cells and K562 cells, scCAPE reveals the heterogeneous perturbation landscapes of T - cell regulators and cell - cycle regulators, and discovers new gene - interactions, such as the synergy between CBL and CNN1 in the S - phase. In conclusion, scCAPE provides a novel and effective method that can analyze the heterogeneous effects of gene perturbations at the single - cell level, providing a powerful tool for in - depth understanding of gene - regulation mechanisms.

Inference of heterogeneous effects in single-cell genetic perturbation screens

Scart: Recognizing Cell Clusters and Constructing Trajectory from Single-Cell Epigenomic Data

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

Linking Genotypes with Multiple Phenotypes in Single-Cell CRISPR Screens

Scmageck Links Genotypes with Multiple Phenotypes in Single-Cell CRISPR Screens.

Predicting Cellular Responses to Complex Perturbations in High-Throughput Screens.

Learning interpretable cellular responses to complex perturbations in high-throughput screens

Scgen Predicts Single-Cell Perturbation Responses

Arbitrary optics - novel nanophotonic and self-adapting optoelectronic systems

Efficient combinatorial targeting of RNA transcripts in single cells with Cas13 RNA Perturb-seq

Robust differential expression testing for single-cell CRISPR screens at low multiplicity of infection

A genome-scale single cell CRISPRi map of trans gene regulation across human pluripotent stem cell lines

High-resolution characterization of gene function using single-cell CRISPR tiling screen

Cobalt‐induced anaphylaxis, contact urticaria, and delayed allergy in a ceramics decorator

Assessment of batch-correction methods for scRNA-seq data with a new test metric

Systematic reconstruction of molecular pathway signatures using scalable single-cell perturbation screens

Identification of cell types, states and programs by learning gene set representations

Decoding Heterogenous Single-cell Perturbation Responses

Dissecting key regulators of transcriptome kinetics through scalable single-cell RNA profiling of pooled CRISPR screens

Higher-order combinatorial chromatin perturbations by engineered CRISPR-Cas12a for functional genomics