Dissecting context-dependent cancer vulnerabilities using Perturb-seq

CRISPR-based viability assays have contributed to cancer research and personalized medicine, but cell fitness is not the only relevant trait. Gene expression analysis reveals cellular stress, inflammation, and altered states, while also revealing activation of cell-death pathways. Perturb-seq, a scalable functional genomics screening method at single-cell resolution, offers the potential to examine a wide range of expression phenotypes. However, current datasets focus on a limited number of widely used cell lines.

In a proof-of-concept study, researchers generated a Perturb-seq dataset targeting 100 genes in 16 different cancer cell lines. During the process, they developed methods to address single-cell technical artifacts, identified Cas9-mediated chromosomal aberrations, and evaluated the quality of the screening. Despite the limited library, the study revealed common patterns when essential genes were deleted, as well as context-specific responses based on the intrinsic genomic properties of the models.

For instance, they inferred a previously unknown connection between dependence on the ER-golgi transport gene immediate early response 3 interacting protein 1 (IER3IP1) and oxidative stress, highlighting the potential of integrated Perturb-seq for hypothesis generation.

The researchers established a framework for creating a comprehensive map of post-perturbational transcriptional phenotypes by conducting parallel Perturb-seq experiments across multiple cell lines. They demonstrated that integrating Perturb-seq experiments across diverse contexts enables the generation of hypotheses about gene function specific to tissue types or cancer subtypes.

The study suggests that large-scale, genome-wide datasets would provide valuable insights into the highly context-dependent nature of cancer biology.