CRISPR/Cas9-based depletion of 16S ribosomal RNA improves library complexity of single-cell RNA-sequencing

Abstract

Single-cell transcriptomics (scRNA-seq) has revolutionized our understanding of cell types and states in various contexts, such as development and disease. To selectively capture protein-coding polyadenylated transcripts, most methodologies rely on poly(A) enrichment to exclude ribosomal transcripts that constitute >80% of the transcriptome. However, it is common for ribosomal transcripts to sneak into the library, which can add significant background by flooding libraries with irrelevant sequences. The challenge of amplifying all RNA transcripts from a single cell has motivated the development of new technologies to optimize retrieval of transcripts of interest. This problem is especially striking in planarians, where a single 16S ribosomal transcript is widely enriched (20-80%) across single-cell methods. Therefore, we adapted the Depletion of Abundant Sequences by Hybridization (DASH) to the standard 10X scRNA-seq protocol. We designed single-guide RNAs tiling the 16S sequence for CRISPR-mediated degradation, and subsequently generated untreated and DASH-treated datasets from the same libraries to enable a side-by-side comparison of the effects of DASH. DASH specifically removes 16S sequences without off-target effects on other genes. By assessing the cell barcodes shared by both libraries, we find that DASH-treated cells have consistently higher complexity given the same amount of reads, which enables the detection of a rare cell cluster and more differentially expressed genes. In conclusion, DASH can be easily integrated into existing sequencing protocols and customized to deplete unwanted transcripts in any organism.