RainDrop: Rapid activation matrix computation for droplet-based single-cell RNA-seq reads

Abstract

Background: Obtaining data from single-cell transcriptomic sequencing allows for the investigation of cell-specific gene expression patterns, which could not be addressed a few years ago. With the advancement of droplet-based protocols the number of studied cells continues to increase rapidly. This establishes the need for software tools for efficient processing of the produced large-scale datasets. We address this need by presenting RainDrop for fast gene-cell count matrix computation from single-cell RNA-seq data produced by 10x Genomics Chromium technology. Results: RainDrop can process single-cell transcriptomic datasets consisting of 784 million reads sequenced from around 8.000 cells in less than 40 minutes on a standard workstation. It significantly outperforms the established Cell Ranger pipeline and the recently introduced Alevin tool in terms of runtime by a maximal (average) speedup of 30.4 (22.6) and 3.5 (2.4), respectively, while keeping high agreements of the generated results. Conclusions: RainDrop is a software tool for highly efficient processing of large-scale droplet-based single-cell RNA-seq datasets on standard workstations written in C++. It is available at https://gitlab.rlp.net/stnieble/raindrop.

Alevin-fry unlocks rapid, accurate and memory-frugal quantification of single-cell RNA-seq data

“…Srivastava et al ( 5 ) introduced , which focused on improving the computational efficiency of tagged-end scRNA-seq quantification and also introduced a novel approach for resolving gene-multimapping UMIs. Likewise, the tool ( 6 ) pairs a custom lightweight mapping approach with a reduced index to count UMIs mapping to genes, providing a fast counting approach. Melsted et al ( 7 ) introduced the pipeline for processing scRNA-seq data; the approach focuses on modularity and speed, using pseudoalignment ( 3 ) to the transcriptome to produce intermediate BUS files ( 8 ) that are subsequently manipulated using commands.…”

Benchmarking of lightweight-mapping based single-cell RNA-seq pre-processing

“…One promising approach to single-cell RNA-seq pre-processing is lightweight-mapping, which was initially developed for bulk RNA-seq in the form of pseudoalignment implemented in the kallisto program (Bray et al 2016) , and later re-implemented in the Salmon program (Patro et al 2017) . More recently, the Salmon pseudoalignment algorithm has been modified to perform "selective alignment" (Srivastava et al 2020) and a sketch alignment strategy was recently implemented in RainDrop, though the tool is limited to processing only a single data type and for that reason we do not consider it here (Niebler et al 2020) . Following the naming convention of (Srivastava et al 2020) we collectively refer to pseudoalignment and selective alignment as "lightweight-mapping" procedures to distinguish them from standard alignment algorithms.…”

A like-for-like comparison of lightweight-mapping pipelines for single-cell RNA-seq data pre-processing

“…Finally, we also note that the benchmark of Booeshaghi and Pachter (1) focuses only on comparing kallisto-bustools to a single configuration of alevin-fry, excluding other relevant tools like STARsolo (9), which is a fast, flexible, and popular tool for the pre-processing of tagged-end single-cell data. The benchmark also omits another recently-published, lightweight-mapping based tool, Raindrop (14), from the benchmark (though, seemingly, this would currently have to be restricted to 10x chromium v2 data). A more extensive benchmark, including other tools, is likely to provide greater value to the broader community.…”