The BIOM file format (canonically pronounced biome) is designed to be a general-use format for representing biological sample by observation contingency tables. BIOM is a recognized standard for the Earth Microbiome Project and is a Genomics Standards Consortium supported project.
The BIOM format is designed for general use in broad areas of comparative -omics. For example, in marker-gene surveys, the primary use of this format is to represent OTU tables: the observations in this case are OTUs and the matrix contains counts corresponding to the number of times each OTU is observed in each sample. With respect to metagenome data, this format would be used to represent metagenome tables: the observations in this case might correspond to SEED subsystems, and the matrix would contain counts corresponding to the number of times each subsystem is observed in each metagenome. Similarly, with respect to genome data, this format may be used to represent a set of genomes: the observations in this case again might correspond to SEED subsystems, and the counts would correspond to the number of times each subsystem is observed in each genome.
There are two components to the BIOM project: first is the definition of the BIOM format, and second is development of support objects in multiple programming languages to support the use of BIOM in diverse bioinformatics applications. The version of the BIOM file format is independent of the version of the biom-format software.
There are official implementations of BIOM format support objects (APIs) in the Python and R programming languages. The rest of this site contains details about the BIOM file format (which is independent of the API) and the Python biom-format API. For more details about the R API, please see the CRAN biom package.
