GENOMICS

SEQUENCING OF GENOMES

As mentioned in Chapter 3, we now know the complete nucleotide sequences of genomes of many organisms. The availability of this large amount of data at an unprecedented scale now allows us, and indeed forces us, to think “globally,” that is, on the scale of whole organisms, or even an assemblage of organisms, rather than of individual genes and enzymes. Here we describe very briefly how the genome sequences are determined.

Genome sequencing of viruses began in the late 1970s. The basic technique involved, the random sequencing of fragments by the Sanger dideoxy termination method, was proposed and applied successfully by Fred Sanger and associates to the complete sequencing of bacteriophage DNAs, notably that of phage λ in 1980 (see Figure 4.1 for the principle of the random shotgun method).

Historically, the first attempt to obtain a complete genome sequence of a cellular organism was geared toward Escherichia coli, the best studied organism outside of humans. This project started in 1989 and used a “directed” approach. Because a fairly detailed genetic map of E. coli was available thanks to the efforts of bacterial geneticists, it was possible to first produce a set of λ-based clones, each containing up to 20 kb DNA, with overlapping ends. The sequencing from here on represented the shotgun phase. The inserted segments in the λ vector were then randomly cut into much smaller fragments of a few kilobases, they were cloned into an M13 vector and were sequenced (for sequencing reactions, see Box 4.1), and the sequences were assembled by looking for overlaps.