Identification of cell barcodes from long-read single-cell RNA-seq with BLAZE

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Long-read single-cell RNA sequencing (scRNA-seq) allows for RNA isoform quantification at the single-cell level.
• Current long-read scRNA-seq protocols, particularly on the Oxford Nanopore platform, often require supplementary short-read data for accurate cell barcode identification.

Purpose of the Study:

• To develop a computational tool, BLAZE, for precise cell barcode identification using only long-read scRNA-seq data.
• To enhance the efficiency and accuracy of long-read scRNA-seq analysis by removing the dependency on matched short-read data.

Main Methods:

• BLAZE employs a novel algorithm to identify 10x cell barcodes directly from nanopore long-read scRNA-seq data.
• Performance evaluation involved comparing BLAZE's cell barcode identification accuracy against existing methods and matched short-read data.

Main Results:

• BLAZE demonstrates superior accuracy and efficiency in identifying cell barcodes compared to existing tools.
• The tool provides a reliable representation of cellular barcodes in long-read scRNA-seq datasets, comparable to results obtained with matched short reads.
• BLAZE successfully simplifies the workflow for long-read scRNA-seq analysis.

Conclusions:

• BLAZE offers a robust and streamlined approach for cell barcode identification in long-read scRNA-seq.
• The tool enhances the utility of nanopore-based long-read scRNA-seq by enabling accurate cell identification without requiring short-read sequencing.
• BLAZE is compatible with standard downstream bioinformatics pipelines and is publicly available.