MicrobiologyBytes

In this article in Microbiology Today Ed Feil describes how we must brace ourselves for the next wave of data as new sequencing techniques become available to determine and compare many sequences at once. The enormous amount of data soon to be generated will bring exciting new insights into how micro-organisms within communities evolve and interact:

Regardless of the species in question, announcements of completed genome sequencing projects in the mainstream media almost invariably make reference to ‘cracking a code’ or ‘deciphering a genetic blueprint’. For bacteria, these over-used analogies spectacularly fail to give a true sense of the fluidity of genome evolution. The doe-eyed assumption in the mid-1990s that a single genome sequence can safely be considered as a prescriptive ‘solution’ for a given bacterial species has been dramatically falsified. By the late 1990s, multiple genome sequences for Escherichia coli revealed extensive differences in gene content between strains, and it rapidly became clear that, for many taxa, an individual genome is most usefully considered as one of many possible combinations of genes drawn from a vast pool known as the pangenome. When faced with such a maelstrom, our natural inclination (as good cladists) is to try and tidy it up, and catalogue strains into pockets of relatedness. Fortunately, phylogenetic analyses are possible, even for very variable species like E. coli, because one can readily identify genes which are universally present in all strains. These essential ‘core’ genes can be thought of as representing the operating system of a given species. In contrast, the specialist software is provided by ‘non-core’ or ‘accessory’ genes which are variably present or absent, are commonly acquired by horizontal transfer, and tend to be restricted to hypervariable regions called genomic islands. These two sets of genes present a fundamental duality in bacterial genomics. Whilst core genes can satisfy our requirements for molecular phylogeny (i.e. what a strain is), accessory genes often play a significant role in adaptation and phenotypic differences (i.e. what a strain does). Conflicts between these two can go a long way to explaining the mystery behind the muddle that is bacterial systematics.

Read more

The concept of species was conceived first by Aristotle ~2,400 years ago, and since then taxonomists of all disciplines have been trying to find those premises that would help to circumscribe the biological units observed in nature. Ever since, the idea behind this term has been a topic of considerable interest that has caused great controversy. Prokaryotes are not exempt from this problem, and even the existence of discrete biological units is being questioned. However, for pragmatic reasons, microbiologists need to deal with a classification of the organisms that they isolate. The ultimate goal of taxonomy is to construct a classification that is of operative and predictive use for any discipline in microbiology and that is also essentially stable. From among the serious classifications, spanning nearly one century, taxonomists have obtained a sound system by classifying prokaryotes based on their phylogenetic, genomic, and phenotypic properties.

The early classification of prokaryotes was based solely on phenotypic similarities, but in the late 1960s some genome-based methods were developed to evaluate genomic interrelationships. Among them, DNA-DNA hybridization (DDH) techniques applied to determine crude genome similarities became popular. DDH tended to reproduce and even improve phenotypically circumscribed organism clusters that were considered to be species. Over the years that followed, the construction of the classification system was based on the fact that DDH could reveal coherent genomic groups (genospecies) of strains generally sharing DDH values with greater than 70% similarity. The comparative study of the different methods, prone to distinct experimental error, indicated that the value of 70% could not be used as absolute boundary, but still a gap between 60 and 70% similarity seemed to embrace clear-cut clusters of organisms. Given the large extent of diversity among prokaryotes, the circumscription of each genospecies would, in addition, be dependent on each group being studied. Nevertheless, the use of DDH has mainly driven the construction of the current prokaryotic taxonomy, as it has become the gold standard for genomically circumscribing species. This parameter has had a similar impact in prokaryotic taxonomy as the interbreeding premise that is the basis for the biological species concept for animal and plant taxonomies. In the late 1980′s, taxonomists already believed that the reference standard for determining taxonomy would be full genome sequences.

DNA-DNA hybridization (DDH) has been used for nearly 50 years as the gold standard for prokaryotic species circumscriptions at the genomic level. It has been the only taxonomic method that offered a numerical and relatively stable species boundary, and its use has had a paramount influence on how the current classification has been constructed. However, now, in the era of genomics, DDH appears to be an outdated method for classification that needs to be substituted. The average nucleotide identity (ANI) between two genomes seems the most promising method since it mirrors DDH closely. This paper examine the work package JSpecies as a user-friendly, biologist-oriented interface to calculate ANI and the correlation of the tetranucleotide signatures between pairwise genomic comparisons. The results agreed with the use of ANI to substitute DDH, with a narrowed boundary that could be set at ≈95–96%. In addition, the JSpecies package implemented the tetranucleotide signature correlation index, an alignment-free parameter that generally correlates with ANI and that can be of help in deciding when a given pair of organisms should be classified in the same species. Moreover, for taxonomic purposes, the analyses can be produced by simply randomly sequencing at least 20% of the genome of the query strains rather than obtaining their full sequence.

Shifting the genomic gold standard for the prokaryotic species definition. PNAS USA October 23 2009. doi: 10.1073/pnas.0906412106

Related:

• What is a bacterial species?
• How many bugs?