Rsubread

Rsubread is a software package written in R that provides fast and accurate alignment of high‑throughput sequencing reads to reference genomes and efficient summarization of read counts for downstream analysis. It includes the command line utilities align, subjunc and featureCounts, as well as an R interface that allows users to perform alignment, splicing junction detection, and feature counting within a single R script. The package was developed by the Computational Genomics Group at the Genome Institute of Singapore and released under an open‑source license.

The aligner uses a seed‑based approach that builds a Burrows‑Wheeler transform of the reference sequence, followed by a suffix‑array based mapping that is robust to sequencing errors and indels. Subjunc extends this mapping to detect splice junctions by aligning split reads and constructing an intron-aware alignment. FeatureCounts implements a fast, multi‑threaded algorithm for counting reads overlapping genomic features such as genes, exons, or transcription factor binding sites, and it supports options for multi‑mapping read handling and strand specificity.

Rsubread is designed to be lightweight and easy to install from CRAN or Bioconductor. It supports a wide range of input formats, including FASTQ, BAM and SAM, and can output SAM/BAM files for further processing. Compatibility with other Bioconductor packages such as GenomicAlignments and Rsamtools facilitates integration into existing pipelines for differential expression, allele‑specific expression, and variant calling.

Benchmarks comparing Rsubread to other aligners such as STAR and TopHat2 show comparable or superior accuracy and faster runtime, particularly for large RNA‑seq datasets. Its efficient memory usage and the ability to perform all steps inside R make it suitable for both small laboratories and large bioinformatics workflows. The current stable release is 3.1.0, with continuous integration and maintenance provided by the Genome Institute of Singapore.