account for biological variation. However, the strain in question
should bear significant similarity to the type strain in order to be
reasonably considered as a member of that species. In order to be
scientifically thorough, this comparison should include the type
strains of all the species within the genus the strain is suspected to
belong to, most importantly the type strain of the type species of
the genus. For a comparison with an even more comprehensive
phylogenic context, the type strains of the type species (and if
possible additional nontype species) from various related genera
should be utilized [62].
Historically, phenotypic features formed the primary basis for
the characteristics analyzed in such comparisons. This was particularly true many decades ago, before Avery identified DNA as the
molecular mechanism of heredity [63]. Recognizing that phenotypes can be quite unreliable, today we rely more heavily (though
not exclusively) on genotypic comparisons. Comparisons of differences between highly conserved genes such as 16S rRNA genes,
hsp65, or rpoB and phylogenetic inferences based upon such alignments have brought a more modern paradigm to species identification. However, it inevitably became clear that in many cases, a
single gene could not provide sufficient resolution to differentiate
species (see Table 2 for examples). Subsequent approaches have
used concatenated gene sequences to improve the phylogenetic
resolution. Because small numbers of genes can still fail to delimit
species, larger sets of genes began to form the basis of analytical
techniques such as multilocus sequence typing (MLST). The logical end point in the progression from one gene to a few to several is
the comparison between whole genomes. Each of these techniques
are discussed in the following sections.
4.1 Phenotypic and
Biochemical Tests
Among the methods commonly used for identification of microorganisms are phenotypic characteristics and the results of biochemical assays. Such methods have been the mainstay of
diagnostic and comparative microbiology for over a century. The
use of such assays for differentiation of the various mycobacterial
species is described in greater detail in preceding chapters.
In theory, with a large enough panel of phenotypic traits, there
should emerge certain characteristics with differences between
some of the species. Typical phenotypes used throughout
microbiology include staining characteristics (e.g., Gram and
acid-fast stains), colony morphology, cellular morphology, motility,
etc. Additionally, organisms can be subjected to a wide variety of
biochemical assays (e.g., catalase, oxidase, Voges-Proskauer, and
carbohydrate metabolism). The results of these tests are generally
reported as positive or negative but may also include weak or
variable reactions. By comparing against the phenotypic combinations that are characteristic to the different species, it should be
possible to characterize a new strain and determine which species
best circumscribes the new strain.
Mycobacterial ID and Characterization Using WGS
419
should bear significant similarity to the type strain in order to be
reasonably considered as a member of that species. In order to be
scientifically thorough, this comparison should include the type
strains of all the species within the genus the strain is suspected to
belong to, most importantly the type strain of the type species of
the genus. For a comparison with an even more comprehensive
phylogenic context, the type strains of the type species (and if
possible additional nontype species) from various related genera
should be utilized [62].
Historically, phenotypic features formed the primary basis for
the characteristics analyzed in such comparisons. This was particularly true many decades ago, before Avery identified DNA as the
molecular mechanism of heredity [63]. Recognizing that phenotypes can be quite unreliable, today we rely more heavily (though
not exclusively) on genotypic comparisons. Comparisons of differences between highly conserved genes such as 16S rRNA genes,
hsp65, or rpoB and phylogenetic inferences based upon such alignments have brought a more modern paradigm to species identification. However, it inevitably became clear that in many cases, a
single gene could not provide sufficient resolution to differentiate
species (see Table 2 for examples). Subsequent approaches have
used concatenated gene sequences to improve the phylogenetic
resolution. Because small numbers of genes can still fail to delimit
species, larger sets of genes began to form the basis of analytical
techniques such as multilocus sequence typing (MLST). The logical end point in the progression from one gene to a few to several is
the comparison between whole genomes. Each of these techniques
are discussed in the following sections.
4.1 Phenotypic and
Biochemical Tests
Among the methods commonly used for identification of microorganisms are phenotypic characteristics and the results of biochemical assays. Such methods have been the mainstay of
diagnostic and comparative microbiology for over a century. The
use of such assays for differentiation of the various mycobacterial
species is described in greater detail in preceding chapters.
In theory, with a large enough panel of phenotypic traits, there
should emerge certain characteristics with differences between
some of the species. Typical phenotypes used throughout
microbiology include staining characteristics (e.g., Gram and
acid-fast stains), colony morphology, cellular morphology, motility,
etc. Additionally, organisms can be subjected to a wide variety of
biochemical assays (e.g., catalase, oxidase, Voges-Proskauer, and
carbohydrate metabolism). The results of these tests are generally
reported as positive or negative but may also include weak or
variable reactions. By comparing against the phenotypic combinations that are characteristic to the different species, it should be
possible to characterize a new strain and determine which species
best circumscribes the new strain.
Mycobacterial ID and Characterization Using WGS
419