[Paper Review] featureCounts: an efficient general-purpose read summarization program

Yang Liao, Gordon K. Smyth+1 more|arXiv (Cornell University)|May 15, 2013

Genomics and Phylogenetic Studies28 citations

TL;DR

featureCounts is a highly efficient read summarization tool designed for counting reads mapping to genomic features such as genes and exons in RNA-seq and whole-genome sequencing data. It uses chromosome hashing and feature blocking to achieve up to tenfold speed improvements and significantly reduced memory usage compared to existing tools, while supporting both single- and paired-end reads with extensive configuration options.

ABSTRACT

Next-generation sequencing technologies generate millions of short sequence reads, which are usually aligned to a reference genome. In many applications, the key information required for downstream analysis is the number of reads mapping to each genomic feature, for example to each exon or each gene. The process of counting reads is called read summarization. Read summarization is required for a great variety of genomic analyses but has so far received relatively little attention in the literature. We present featureCounts, a read summarization program suitable for counting reads generated from either RNA or genomic DNA sequencing experiments. featureCounts implements highly efficient chromosome hashing and feature blocking techniques. It is considerably faster than existing methods (by an order of magnitude for gene-level summarization) and requires far less computer memory. It works with either single or paired-end reads and provides a wide range of options appropriate for different sequencing applications. featureCounts is available under GNU General Public License as part of the Subread (this http URL) or Rsubread (this http URL) software packages.

Motivation & Objective

• To address the lack of attention given to read summarization in genomics despite its central role in downstream analysis.
• To develop a faster and more memory-efficient method for counting reads mapped to genomic features such as genes and exons.
• To support diverse sequencing applications by providing flexible options for single- and paired-end reads.
• To improve scalability and performance for large-scale genomic datasets through optimized data structures and algorithms.

Proposed method

• Implementation of chromosome hashing to accelerate the lookup of read alignments to genomic features.
• Use of feature blocking to partition the genome into manageable blocks, reducing computational overhead.
• Design of a multi-threaded architecture to leverage modern multi-core processors for parallel processing.
• Integration of support for both single-end and paired-end sequencing reads with configurable alignment filtering.
• Optimization of memory access patterns to minimize cache misses and improve CPU efficiency.
• Incorporation of configurable parameters for sensitivity, alignment filtering, and output formats to suit various experimental designs.

Experimental results

Research questions

• RQ1 Can a read summarization tool achieve significantly improved performance over existing methods in terms of speed and memory usage?
• RQ2 How does chromosome hashing and feature blocking affect the efficiency of read-to-feature assignment in large-scale sequencing data?
• RQ3 To what extent can featureCounts scale across different sequencing depths and feature types (e.g., genes, exons)?
• RQ4 Does the tool maintain accuracy and flexibility while reducing computational overhead?

Key findings

• featureCounts achieves up to a tenfold speedup in gene-level read summarization compared to existing tools.
• The program requires substantially less computer memory than prior methods, enabling efficient processing on standard hardware.
• Chromosome hashing and feature blocking techniques significantly reduce I/O and computational bottlenecks during read counting.
• The tool supports both single-end and paired-end reads with high configurability for diverse sequencing applications.
• featureCounts maintains high accuracy and consistency across various genomic feature types and sequencing depths.