Comparison of High-Throughput Single-Cell RNA Sequencing Data Processing Pipelines

Abstract

With the development of single-cell RNA sequencing (scRNA-seq) technology, it has become possible to perform large-scale transcript profiling for tens of thousands of cells in a single experiment. Many analysis pipelines have been developed for data generated from different high-throughput scRNA-seq platforms, bringing a new challenge to users to choose a proper workflow that is efficient, robust and reliable for a specific sequencing platform. Moreover, as the amount of public scRNA-seq data has increased rapidly, integrated analysis of scRNAseq data from different sources has become increasingly popular. However, it remains unclear whether such integrated analysis would be biased if the data were processed by different upstream pipelines. In this study, we encapsulated seven existing high-throughput scRNA-seq data processing pipelines with Nextflow, a general integrative workflow management framework, and evaluated their performances in terms of running time, computational resource consumption, and data processing consistency using nine public datasets generated from five different high-throughput scRNA-seq platforms. Our work provides a useful guideline for the selection of scRNA-seq data processing pipelines based on their performances on different real datasets. In addition, these guidelines can serve as a performance evaluation framework for future developments in high-throughput scRNA-seq data processing.Since the emergence of the first single-cell RNA sequencing (scRNA-seq) platform [1], many research achievements have been made at the cellular and subcellular levels with unprecedented resolutions with the aid of this technology. Recent advances in microfluidics and next generation sequencing (NGS) have further increased the efficiency and throughput of scRNA-seq, enabling more cells to be identified and the expression information of more genes for each cell to be quantified simultaneously [2,3,4]. From microcapillary pipettes in 2012 [5,6] and nanoliter droplet-based microfluidic chips in 2015 [7,8] to the latest liquid barcoding combined with split-pool methods [9], the number of cells analysed in parallel has increased from several to tens of thousands [10], bringing new challenges for the processing of a large amount of barcoded NGS data for efficient and accurate quantification of the transcript information of single cells.To meet the needs of high-throughput scRNA-seq data processing, several pipelines that integrate multiple functions have been developed. As one of the first high-throughput scRNA-seq platforms, Drop-seq [7] was introduced in 2015. Moreover, Drop-seq-tools in combination with Picard tools were introduced and provided a user-friendly Application Programming Interface (API) to process the data from Drop-seq. Later in 2017, three additional pipelines were published including Cell Ranger [11], umis [12] and UMItools [13]. Cell Ranger [11] was developed along with the widely adopted 10X Genomics platform which can process multiple biological samples separated by sample barcodes in paral...

Eoulsan 2: an efficient workflow manager for reproducible bulk, long-read and single-cell transcriptomics analyses

“…Eoulsan relies on state-of-the-art tools such as UMI-Tools, which enables more versatility than CellRanger. CellRanger's lack of time efficiency and high requirement for memory usage has previously been underlined (Gao et al 2021). Our workflow is 2-3 times faster with limited resource usage for similar results (R2=0.998 for UMI count per cell) (Figure 4).…”

From bench to bedside: Single-cell analysis for cancer immunotherapy

“…The preprocessing of FASTQ files involves alignment to the human transcriptome and quantification of reads for each transcript or UMI, depending on the technology. Some software also performs quality control while pre-processing the raw data (Gao et al, 2020). These alignment and pre-processing steps return a matrix containing barcodes specific to each cell captured and the counts for detected transcripts.…”

pipeComp, a general framework for the evaluation of computational pipelines, reveals performant single cell RNA-seq preprocessing tools

“…A number of good comparison and benchmark studies have already been performed on various steps related to scRNAseq processing and analysis and can guide the choice of methodology [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21]. However, these recommendations need constant updating and often leave open many details of an analysis.…”