rseqc

TAFFISH wrapper for RSeQC 5.0.4, a Python command suite for RNA-seq quality-control metrics from SAM/BAM alignments and BED/GTF annotations.

Package identity:

• name: rseqc
• command: taf-rseqc
• kind: tool
• TAFFISH version: 5.0.4-r2
• container image: ghcr.io/taffish/rseqc:5.0.4-r2
• upstream package: RSeQC 5.0.4 from PyPI
• default upstream command: bam_stat.py
• supported platforms: linux/amd64, linux/arm64

Installation

taf install rseqc

or run the built wrapper directly during development:

taf build
TAFFISH_CONTAINER_BACKEND=docker target/taf-rseqc-v5.0.4-r2 bam_stat.py -i sample.bam

Usage

RSeQC is a suite of scripts rather than a single rseqc executable. The wrapper keeps bam_stat.py as the default command so option-leading calls still work:

taf-rseqc -- --version
taf-rseqc -- -i sample.bam

For normal work, prefer command mode and name the RSeQC script explicitly:

taf-rseqc bam_stat.py -i sample.bam
taf-rseqc infer_experiment.py -i sample.bam -r genes.bed
taf-rseqc read_distribution.py -i sample.bam -r genes.bed
taf-rseqc geneBody_coverage.py -i sample.bam -r genes.bed -o sample.gene_body
taf-rseqc tin.py -i sample.bam -r genes.bed
taf-rseqc sc_seqLogo.py -i barcodes.fq.gz --iformat fq --oformat png -o barcodes

taf-rseqc --help and taf-rseqc --version are TAFFISH wrapper-level options. Use taf-rseqc bam_stat.py --help, taf-rseqc infer_experiment.py --help, or taf-rseqc -- --help for upstream help.

Packaged commands

The image installs all scripts shipped by the upstream RSeQC 5.0.4 source distribution:

• alignment and coverage: bam_stat.py, bam2fq.py, bam2wig.py, divide_bam.py, split_bam.py, split_paired_bam.py
• RNA-seq annotation checks: infer_experiment.py, read_distribution.py, geneBody_coverage.py, geneBody_coverage2.py, tin.py
• read-level QC: read_GC.py, read_NVC.py, read_quality.py, read_duplication.py, read_hexamer.py, clipping_profile.py, deletion_profile.py, insertion_profile.py, mismatch_profile.py
• saturation and fragments: RPKM_saturation.py, RNA_fragment_size.py, inner_distance.py, junction_annotation.py, junction_saturation.py
• quantification helpers: FPKM_count.py, FPKM-UQ.py
• BigWig utilities: normalize_bigwig.py, overlay_bigwig.py
• single-cell helpers: sc_bamStat.py, sc_editMatrix.py, sc_seqLogo.py, sc_seqQual.py

The container also includes Rscript for RSeQC plot-generation paths, htseq-count for FPKM-UQ.py, GNU awk for single-cell summary helpers, and Python dependencies including pysam, bx-python, numpy, pyBigWig, matplotlib, and logomaker. numpy is kept on the 1.x ABI line so the Debian pyBigWig and HTSeq helper packages remain compatible on both amd64 and arm64.

The upstream documentation copied into the image is normalized to root-owned files so rootless Podman can pull the image without requiring unusual subuid/subgid mappings.

Inputs and outputs

Most RSeQC modules consume sorted SAM/BAM files. BAM-based commands commonly expect a .bai index next to the BAM file. Annotation-aware modules usually require BED12 gene models; FPKM-UQ.py requires a GTF plus a TCGA-style gene information table. Keep BAM, BED/GTF, and output paths under the current working directory or otherwise mounted paths so the container can see them.

Many RSeQC commands write text tables plus an .r plotting script and a PDF/PNG/JPEG plot. The image sets MPLBACKEND=Agg for headless Matplotlib use and includes base R for the R plotting scripts.

Boundaries

This app packages the RSeQC command suite and its direct runtime helpers. It does not bundle RNA-seq aligners, reference genomes, gene-model databases, Cell Ranger, samtools, UCSC browser data, or large annotation downloads. Users must provide the appropriate BAM/SAM, BED/GTF, chromosome-size, and gene-information files.

bam2wig.py can write WIG output and upstream code can try to call UCSC wigToBigWig for BigWig conversion. wigToBigWig is not bundled here because the common upstream binary is platform-specific; use the separate ucsc-kent app if you need a reproducible BigWig conversion step.

RSeQC upstream metadata is slightly inconsistent: PyPI classifiers list GPLv2, while the 5.0.4 source LICENSE file and SourceForge documentation state GPL version 3 or later. This TAFFISH app records the upstream license as GPL-3.0-or-later because that is what the shipped source license says.

Upstream

• Homepage and manual: https://rseqc.sourceforge.net/
• PyPI release: https://pypi.org/project/RSeQC/5.0.4/
• Source archive used here: https://files.pythonhosted.org/packages/f8/04/f32be86c3e14fa77b0dae3a9cdea9e5795642960ade79e3d463664101a4d/rseqc-5.0.4.tar.gz
• Source SHA256: b7f3996f3de0b0b0a09eec949281a8f3e665a20827fcb3cbbd7546b94574a088
• Citation: Wang, L., Wang, S., and Li, W. 2012. RSeQC: quality control of RNA-seq experiments. Bioinformatics. DOI: 10.1093/bioinformatics/bts356; PMID: 22743226.
• TIN citation: Wang et al. 2016. Measure transcript integrity using RNA-seq data. BMC Bioinformatics. DOI: 10.1186/s12859-016-0922-z.