R in a Nutshell, 2nd Edition

Data Structures

One of the best features of Bioconductor is the use of structured data to represent biological concepts. This section presents a few important classes that are used through Bioconductor.

Bioconductor classes are implemented using formal class methods; see Chapter 10 for more details. Most of these classes inherit from the basic classes in the Biobase package, so you can use the same methods to work with different types of objects. For example, you can use the same method to read phenotype data for expression data from different vendors (such as Affymetrix arrays and Illumina arrays). You could also use the same method to read phenotype data for expression data from completely different types of data (such as gene expression data and proteomic data).

Objects in Bioconductor contain many different types of information about an experiment: the experimental platform, information about the samples, information about the phenotypes, the experimental results, and almost anything else that is relevant for describing an experiment or the results of the experiment. Classes defined in the Biobase package provide a general framework that fits many different types of experimental data. Classes defined in other packages can be used to represent data from specific types of microarrays, often for specific products from specific vendors. This section contains descriptions of a few key classes defined in Biobase.

eSet

eSet is a virtual class that is used by many Bioconductor functions. ...