DGEList

DGEList is a data container used in the edgeR package of Bioconductor to store and organize count-based gene expression data for differential expression analysis. It is designed to hold the essential components of digital gene expression experiments, such as RNA-seq, and to be the input for downstream normalization, dispersion estimation, and statistical testing.

A DGEList object is a list-like S3 class that typically includes several components. The counts element is a matrix of integer read counts with genes as rows and samples as columns. The genes element is an optional data frame containing gene annotations, such as gene identifiers and symbols. The samples element is an optional data frame with sample metadata, including group assignment, batch information, and other factors. Additional fields commonly stored in a DGEList are lib.size, a numeric vector of library sizes for each sample, and norm.factors, a numeric vector of normalization factors produced by methods such as TMM normalization. The group information is often stored within the samples data frame and is used to define experimental comparisons.

Creation of a DGEList typically involves the DGEList() constructor, with counts and optional genes or group information supplied by the user. Once created, the object serves as input to core edgeR workflows, including normalization, dispersion estimation, and fitting generalized linear models for differential expression testing. The design of the object emphasizes interoperability with edgeR functions, which operate on DGEList objects and update its components as analyses proceed.

Users can subset or modify a DGEList using standard R operations and dedicated accessor functions to retrieve or update components such as counts, genes, samples, lib.size, and norm.factors. The class provides a structured, consolidated way to manage the data and annotations required for rigorous differential expression analysis within edgeR.