rhapsodist: a reproducible Snakemake workflow for BD Rhapsody single-cell RNA-seq data

preprint OA: closed
Full text JSON View at publisher

Abstract

BD Rhapsody is a widely used single-cell RNA-seq (scRNA-seq) platform to profile gene expression. Based on barcoded beads, processing BD Rhapsody experiments is demanding because of the multiple barcode versions and layouts, which makes it difficult to locate barcodes by scanning reads positionally. We introduce rhapsodist , a reproducible and scalable Snakemake workflow to process BD Rhapsody whole transcriptome analysis (WTA) data from raw FASTQ reads. As a result, rhapsodist generates count tables in a HDF5-backed SingleCellExperiment format. It supports all generations of BD Rhapsody beads, regardless of the barcode allowlist version and the presence of a variable-length diversity inset. rhapsodist orchestrates STARsolo, kallisto/bustools, salmon/alevin, and the official BD Rhapsody CWL pipeline, enabling direct cross-aligner comparison. Optionally, sample tag demultiplexing, read downsampling, tunable linker-mismatch tolerance, and per-sample reports are provided. We showcase a remarkable consistency across aligner results, and highlight their differences in performance (e.g., quantification, clustering, or speed). On simulated data all four aligners recovered true barcodes with perfect precision and recall and equivalent UMI counts. On HeLa cells (a cell line; enhanced beads), both per-cell correlations across aligners and pseudobulk correlations were remarkably high on every comparison (Pearson r of 0.89–0.96 and 0.94–0.98 per cell and pseudobulk, respectively). On a mouse epidermis dataset with a more complex transcriptional profile (e.g., with celltypes; on legacy beads), cell type pseudobulk correlations showed a similar consistency (Pearson r 0.92–0.98), pointing to calling robustness. As expected, pseudoaligners (kallisto/bustools and salmon/alevin) were best performers speed-wise, and BD’s workflow recovered a higher number of barcodes and UMIs. rhapsodist is available at https://github.com/imallona/rhapsodist under GPLv3, tried on Linux with conda, and includes unit and integration tests.
Full text 2,184 characters · extracted from oa-doi-fallback · click to expand
Abstract BD Rhapsody is a widely used single-cell RNA-seq (scRNA-seq) platform to profile gene expression. Based on barcoded beads, processing BD Rhapsody experiments is demanding because of the multiple barcode versions and layouts, which makes it difficult to locate barcodes by scanning reads positionally. We introduce rhapsodist, a reproducible and scalable Snakemake workflow to process BD Rhapsody whole transcriptome analysis (WTA) data from raw FASTQ reads. As a result, rhapsodist generates count tables in a HDF5-backed SingleCellExperiment format. It supports all generations of BD Rhapsody beads, regardless of the barcode allowlist version and the presence of a variable-length diversity inset. rhapsodist orchestrates STARsolo, kallisto/bustools, salmon/alevin, and the official BD Rhapsody CWL pipeline, enabling direct cross-aligner comparison. Optionally, sample tag demultiplexing, read downsampling, tunable linker-mismatch tolerance, and per-sample reports are provided. We showcase a remarkable consistency across aligner results, and highlight their differences in performance (e.g., quantification, clustering, or speed). On simulated data all four aligners recovered true barcodes with perfect precision and recall and equivalent UMI counts. On HeLa cells (a cell line; enhanced beads), both per-cell correlations across aligners and pseudobulk correlations were remarkably high on every comparison (Pearson r of 0.89–0.96 and 0.94–0.98 per cell and pseudobulk, respectively). On a mouse epidermis dataset with a more complex transcriptional profile (e.g., with celltypes; on legacy beads), cell type pseudobulk correlations showed a similar consistency (Pearson r 0.92–0.98), pointing to calling robustness. As expected, pseudoaligners (kallisto/bustools and salmon/alevin) were best performers speed-wise, and BD’s workflow recovered a higher number of barcodes and UMIs. rhapsodist is available at https://github.com/imallona/rhapsodist under GPLv3, tried on Linux with conda, and includes unit and integration tests. Competing Interest Statement GM has received free or discounted kits from BD for an unrelated project. No other competing interests exist.

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: oa-doi-fallback

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2026) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-06-13T06:42:57.164913+00:00