edgeR: a Bioconductor package for differential expression analysis of digital gene expression data.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Microarray technology is being superseded by digital gene expression (DGE) technologies for functional genomics.
• Analyzing gene expression data requires methods to identify significant differences in transcript or exon counts across experimental conditions.
• The Bioconductor software package edgeR is designed for analyzing replicated count data.

Purpose of the Study:

• To introduce edgeR, a software package for differential gene expression analysis.
• To describe the statistical methodology implemented in edgeR for analyzing count data.
• To highlight the applicability of edgeR in various functional genomics and other count-based data analyses.

Main Methods:

• Utilizes an overdispersed Poisson model to capture biological and technical variability in count data.
• Employs empirical Bayes methods to moderate overdispersion across transcripts, enhancing inference reliability.
• Designed to handle count data from digital gene expression experiments, including those with minimal replication.

Main Results:

• Provides a robust method for detecting differential gene expression from replicated count data.
• Improves the reliability of statistical inference by moderating overdispersion.
• Demonstrates utility even with limited experimental replication.

Conclusions:

• Emerging DGE technologies are poised to replace microarrays in many functional genomics applications.
• edgeR offers a reliable and flexible tool for differential expression analysis of count data.
• The methodology has potential applications beyond sequencing, including proteomic peptide count data.