nf-core pipeline · 10x-genomics, 10xgenomics, alevin
nf-core/scrnaseq
Processes single-cell RNA sequencing data across multiple aligners into count matrices for transcriptomic research.
What nf-core/scrnaseq is
nf-core/scrnaseq is a bioinformatics analysis pipeline designed for processing single-cell RNA sequencing (scRNA-seq) data from barcode-based protocols such as 10x Genomics, DropSeq, and SmartSeq. Originally written by P. J. Bailey, O. Botvinnik, F. Marques de Almeida, G. Gabernet, A. Peltzer, and G. Sturm, the workflow automates alignment, barcode processing, and matrix generation.
Analysing droplet single-cell libraries requires mapping sequenced reads, demultiplexing cellular barcodes, resolving unique molecular identifiers (UMIs), and distinguishing genuine cells from empty droplets. The pipeline unifies multiple independent alignment and pseudoalignment tools into a standardised workflow to yield processed count matrices suitable for statistical analysis.
What the pipeline does
The pipeline ingests raw FASTQ files defined in a CSV samplesheet alongside a reference genome FASTA and GTF annotation file. Users specify the library protocol (such as 10XV2) and choose an alignment workflow: SimpleAF (Alevin-Fry) with qcatch, STARsolo, Kallisto with BUStools (kbpython), or CellRanger.
Following read alignment, demultiplexing, and empty-droplet detection, the pipeline outputs processed cellular gene expression count matrices. These results are formatted directly into Matrix Market (mtx), AnnData (h5ad), and Seurat compatible files.
Key facts
- Pipeline framework
- Nextflow
- Latest release
- 4.2.0 (2026-07-04)
- Supported aligners
- STARsolo, Alevin-Fry (SimpleAF), Kallisto + BUStools (kbpython), CellRanger
- Supported protocols
- 10x Genomics, DropSeq, SmartSeq
- Input formats
- FASTQ, FASTA, GTF, samplesheet CSV
- Output formats
- h5ad, Seurat, mtx
- Original authors
- Bailey PJ, Botvinnik O, Marques de Almeida F, Gabernet G, Peltzer A, Sturm G
What you can use it for
- 10x Genomics quantification
- Processing raw single-cell RNA-seq libraries generated from standard 10x Genomics droplet kits into cellular expression matrices.
- Cross-aligner benchmarking
- Comparing quantification outputs and cell-calling results across STARsolo, Alevin-Fry, Kallisto/BUStools, and CellRanger on identical datasets.
- DropSeq and SmartSeq processing
- Aligning and demultiplexing reads from barcode-based protocols beyond standard 10x platforms.
- Downstream analysis preparation
- Producing pre-filtered AnnData (h5ad) or Seurat objects for single-cell clustering, trajectory inference, and differential expression tools.
Run it here
The research agent runs nf-core/scrnaseq with Nextflow on its own machine — up to 32 cores, 256 GiB of memory and 24 hours per run, which covers the test profile and real single-sample runs. It builds the samplesheet, checks it against the pipeline's schema and reads the results back. When a run is too big for one machine, it gives you the exact command for your own cluster instead.
Strengths and limits
Good at
- Supports four different alignment and quantification engines: Alevin-Fry, STARsolo, Kallisto/BUStools, and CellRanger.
- Generates output count matrices directly in standard AnnData (h5ad), Seurat, and Matrix Market (mtx) formats.
- Incorporates empty-droplet detection to distinguish intact cells from ambient RNA background.
- Handles multiple barcode-based library protocols including 10x Genomics, DropSeq, and SmartSeq.
Watch out for
- CellRanger execution relies on a community-built container that is not directly supported by 10x Genomics and falls under its proprietary end-user licence agreement.
- Multi-ome single-cell analysis types are not supported in the pipeline.
- Pipeline configuration files cannot be used to set workflow parameters, which must be passed via the command line or parameter files.
Questions
Can I run nf-core/scrnaseq without my own cluster?
The research agent runs nf-core/scrnaseq with Nextflow on its own machine — up to 32 cores, 256 GiB of memory and 24 hours per run, which covers the test profile and real single-sample runs. It builds the samplesheet, checks it against the pipeline's schema and reads the results back. When a run is too big for one machine, it gives you the exact command for your own cluster instead.
What licence is nf-core/scrnaseq under?
MIT, like every nf-core pipeline.
Which aligners can be selected in the workflow?
The pipeline supports four alignment choices: SimpleAF (Alevin-Fry) paired with qcatch, STARsolo, Kallisto combined with BUStools (kbpython), and 10x Genomics CellRanger.
What inputs does the pipeline require?
Users must supply a CSV samplesheet listing sample names, FASTQ paths, and expected cell counts, along with a reference genome FASTA, a GTF annotation file, and the library protocol.
What matrix formats are generated at the end of the run?
The pipeline exports quantified single-cell gene expression data as AnnData (h5ad), Seurat, and Matrix Market (mtx) files.
Can protocols other than 10x Genomics be analysed?
Yes, the pipeline supports other barcode-based single-cell protocols, such as DropSeq and SmartSeq.
How to cite it
doi: 10.5281/zenodo.3568187
Written from nf-core/scrnaseq's own README. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.