Transcriptomics · Software
featureCounts
Assigns mapped sequencing reads to genomic features such as genes, exons, and promoters for expression analysis.
What featureCounts is
featureCounts is a quantification tool designed to assign mapped sequencing reads to defined genomic features, including genes, exons, promoters, and genomic bins. It was created as part of the Subread package and is also available within the Bioconductor package Rsubread.
The software addresses the need to quantify transcript levels and genomic coverage for transcriptomic workflows, such as RNA-seq expression analysis. It processes read alignments against genomic annotations, summarising counts across single-end, paired-end, or long-read sequencing datasets.
How it works
featureCounts takes aligned sequencing reads in SAM or BAM format and feature coordinates specified in Simplified Annotation Format (SAF) or GTF/GFF format, with support for gzipped annotation files and inbuilt reference sets like mm39. Users specify target feature types, and the tool evaluates overlapping alignments against these features.
The assignment process can be customised using various rules. These include configuring stranded or unstranded counting per library, counting whole read pairs or individual reads, requiring minimum fractional overlap thresholds, or allocating reads by largest overlap. In addition to feature counting, it can extract and tally reads spanning exon-exon junctions and separate counts by read groups.
Key facts
- Latest version
- 2.1.1
- Software suite
- Subread (also implemented in Rsubread)
- Primary domain
- Transcriptomics
- Input formats
- SAM, BAM
- Annotation formats
- SAF, GTF, GFF (supports gzipped files)
- Supported genomic targets
- Genes, exons, promoters, genomic bins, and exon-exon junctions
What you can use it for
- RNA-seq gene and exon quantification
- Count aligned sequence reads mapping to genes, exons, or user-defined genomic bins to measure expression levels.
- Exon-exon junction discovery and counting
- Quantify reads that span exon-exon junctions in both short-read and long-read data using GTF or SAF annotations.
- Read group-level quantification
- Segregate read counts by read group identifiers specified in SAM or BAM headers.
- Paired-end fragment counting
- Quantify read pairs directly as single fragments rather than treating paired mates as independent reads.
Run it here
featureCounts is preinstalled on the machine behind every research conversation. Ask the agent in plain language, or give it the exact command — it runs it, reads the output and explains it.
Try asking
“Count reads per gene in these BAM files with featureCounts using this GTF.”
Strengths and limits
Good at
- Accepts BAM and SAM files containing a mixture of single-end and paired-end reads.
- Provides fine-grained overlap controls, including fractional overlap filtering and largest-overlap assignment.
- Reports exon-exon junction counts from annotated splice junctions.
- Supports multiple annotation formats including SAF, GTF, GFF, and inbuilt genome annotations.
Watch out for
- Requires pre-aligned SAM or BAM inputs rather than raw FASTQ sequence files.
- The '--maxMOp' parameter has been deprecated and is no longer available for alignment operation parsing.
Questions
Can I run featureCounts online without installing it?
featureCounts is preinstalled on the machine behind every research conversation. Ask the agent in plain language, or give it the exact command — it runs it, reads the output and explains it.
What licence is featureCounts under?
featureCounts is published under its authors' own licence; the project page at https://subread.sourceforge.net/ states it.
Which annotation formats can be used with featureCounts?
featureCounts accepts Simplified Annotation Format (SAF) files as well as GTF and GFF annotations, which can also be provided as gzipped files.
Can featureCounts handle paired-end sequencing data?
Yes. Paired-end reads are indicated using the '-p' flag, and read pairs can be tallied explicitly as fragments using the '--countReadPairs' option.
Does featureCounts support strand-specific RNA-seq?
Yes. Stranded or unstranded counting can be configured individually for each library via the '-s' parameter.
Is featureCounts accessible in R?
Yes. The program is implemented within the R ecosystem as part of the Bioconductor package Rsubread.
How to cite it
Liao Y, Smyth GK and Shi W. featureCounts: an efficient general-purpose program for assigning sequence reads to genomic features. Bioinformatics , 30(7):923-30, 2014
Written from featureCounts's own README and project page. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.