Figure 1. Conceptualizing the Genome.

The determination of many thousands of genomes will require a precise definition of what a genome sequence represents. We envisage a hierarchy in which a reference genome comprising all known sequences within a species is placed at the topmost level. A subset of the reference sequence defines a strain or population genome that includes all known polymorphisms within the population. At the individual level is a uniquely defined genome. Such a defined hierarchy would facilitate unique identifiers for classes within each level (for example, for a microbial isolate a Unique Genome Identifier could take the form XXX-YYY-ZZZ, where XXX denotes the species, YYY denotes the strain, and ZZZ denotes the specific isolate sequenced). This hierarchy would also enable efficient data storage of complete genome information for individuals, because the information stored at a lower level of specification needs only to describe what is specific to that level.