17.8 Metagenomics

267

organisms, from bacteria to man. Later, the sequence of the small subunit of ribosomal

RNA (rRNA), another essential and universal object, was used. ²⁷ Nowadays, one can,

in principle, analyse whole genomes.

A chronology can be established on the premiss that the more changes there are, the

longer the time elapsed since the species diverged (assuming that the changes occur

at a constant rate with respect to sidereal time). This premiss can be criticized since,

although the unit of change is the nucleotide, selection (the engine of speciation) acts

on the amino acid; some nucleotide mutations lead to no change in amino acid due

to the degeneracy of the code. There is actually little real evidence that mutations

occur at random (i.e., with respect to both the site and the type of mutation).

A difficulty with molecular phylogenies is the fact that lateral gene transfer (LGT;

cf. Sect. 14.7.4), especially between bacteria and between archaea, may vitiate the

calculated distances. A plausible counterargument in favour of the use of rRNA is

that it should be unaffected by LGT, due to its fundamental place in cell metabolism.

A further difficulty is a computational one: that of finding the optimal tree, since

usually one is interested in comparing dozens (and ultimately millions) of species.

The basic principle applied to address this problem is that of parsimony: One seeks to

construct the tree with the least possible number of evolutionary steps. Unfortunately,

this is an NP-complete problem and hence the computation time grows exponentially

with the number of species; even a mere 20 species demands the analysis of almost

10 Superscript 2210²² possible trees!

17.8

Metagenomics

The ability to culture bacteria in the laboratory, of which Pasteur seems to have been

the pioneer, was a crucial step in the emergence of bacteriology. Pasteur used liquid

media (broths); for the purposes of investigation, however, solid media, introduced

by Koch, are more convenient. Since then, the culture of bacteria has become indis-

pensable to vast areas of medicine, biotechnology, and research, for the identification

and counting of bacteria and for the development of serological assays and vaccines,

to name just a few ways of making use of bacterial cultures. At the same time, it

is recognized that the vast majority of bacteria cannot be cultured. Therefore, the

only way that this extant, almost immanent microbial richness, can be accessed is

by sequencing its genomic signature. The genomes of the entire natural microbiota

collectively constitute the metagenome. ²⁸

The vast increase in DNA sequencing capability—both in terms of hardware and

in algorithms for analysing the raw data—has allowed metagenomics (the study of the

metagenome) to become a practical science. Work begins with the extraction of the

DNA of all the microbes in some environmental sample (e.g., soil, ²⁹ or seawater, or

27 rRNA has been championed by C. Woese (2000).

28 Rondon et al. (2000); Committee on Metagenomics (2007).

29 Ritz (2008).