Non-variable RNA deletion using the CRISPR-Cas9 technique demonstrated improved outcomes in human intestine single-cell RNA sequencing data, even at half sequencing depths

TL;DR

A CRISPR-Cas9 method to remove non-variable RNAs in human intestinal scRNA-seq data improves quality and reduces costs without harming biological integrity.

Contribution

A novel CRISPR-Cas9-based approach for RNA deletion that outperforms computational methods in scRNA-seq data quality.

Findings

CRISPR-Cas9 treatment effectively reduced non-variable RNA expression, including ribosomal and mitochondrial RNAs.

Treated samples at half sequencing depth maintained comparable quality and saturation to untreated samples.

Cell type composition and gene expression patterns remained consistent after treatment with no unintended deletions.

Abstract

In single-cell RNA sequencing (scRNA-seq) data, issues related to the high expression of non-variable RNAs often arise due to organ traits or sample quality. Computational methods, such as SoupX (Young (Gigascience 9:giaa151, 2020)), have been used to solve this problem but it may remove biologically relevant data. This study presents a clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9-based method that selectively removes non-variable RNAs. We applied this approach to scRNA-seq data from human intestinal tissues of 17 patients. By targeting non-variable genes, including ribosomal and mitochondrial RNAs, CRISPR-Cas9 treatment effectively reduced their expression, outperforming computational methods in both the number and extent of gene removal. The CRISPR-Cas9 treated samples, sequenced at half the depth compared to untreated samples, maintained comparable sequencing quality, and saturation, demonstrating that this approach can reduce sequencing costs while preserving data quality. Cell type composition and gene expression patterns remained consistent between treated and original datasets, with no unintended gene deletions. Overall, our findings suggest that the CRISPR-Cas9-based method offers a cost-effective solution for improving scRNA-seq data quality, particularly for tissues with high levels of non-variable RNAs, without compromising biological integrity.

The online version contains supplementary material available at 10.1186/s44342-025-00043-6.

Figures