LAUSR

In the beginning was... the genome. The 30th of January, 1991 found me outside Temple Underground Station on the Thames Embankment in London. John Sgouros of the Martinsried Institute for Protein Sciences, who had flown over from Munich that morning, emerged from the station carrying seven A0 sheets in a roll. They were copies of Figure 1 for the paper ‘The complete DNA sequence of yeast chromosome III’, for which we were determined to get a 1991 submission date. We walked around the corner to the Nature offices in Little Essex Street and handed in these giant Figures together with envelopes containing the manuscript and the other Figures, which I had brought down from Manchester. Nature then lost the copies of Figure 1 but, by 27th March 1992, the paper was accepted and appeared in the journal on the 7th of May 1992 (Oliver et al. 1992). This paper reported the first complete DNA sequence of a chromosome from any organism and, with it, everything changed— certainly for me, for yeast genetics, but also for the wider biomedical research community’s view of the value of genome sequencing. The complete sequence of this chromosome taught us some important lessons in two major areas. First, in the area of eukaryotic chromosome organisation and evolution. We found that the relationship between the genetic distance between loci, as measured in terms of recombination frequency, and that of physical distance, as measured by the DNA sequence, was far from linear. In fact, the ratio of the genetic distance (in cM) to the physical distance (in kb) showed a > 40fold range for different intervals along the chromosome, being smallest close to the centromere and greatest half-way down each chromosome arm (Figure 3 in Oliver et al. 1992, 1993). It also indicated the role of retrotransposons in the generation of redundancy in the yeast genome by duplicating portions of the chromosome (Wicksteed et al. 1994) and increasing the copy number of tRNA genes (Eigel and Feldmann 1982, Genbauffe et al. 1984). Redundancy is the substrate exploited by evolution to generate novel functions (Ohno 1970, and see below). Second, the complete sequence of this chromosome taught us that we knew much less about the genetics of Saccharomyces cerevisiae than we thought. The yeast genetics research community, at this time, was justly proud of the genome map (Mortimer et al. 1989) that had been constructed by classical genetic techniques supplemented by recombinant DNA analyses. In fact, there was even talk that for some functions, e.g. DNA repair, the map was saturated. However, inspection of the chromosome III sequence showed that there were 182 potential protein-encoding genes (open reading frames, ORFs, > 100 amino acids in length) of which only 37 were known from classical studies (Oliver et al. 1992). There were 35 laboratories involved in the sequencing of this first chromosome, and the Yeast Genome Consortium, ably led by André Goffeau, went on in this ‘cottage industry’ approach to sequence the other 15 chromosomes until the first complete genome sequence of a eukaryotic organism was published in 1996 (Goffeau et al. 1996), just a year after that of the first complete bacterial genome sequence (Fleischmann et al. 1995). There were individual papers reporting the sequence of the other 15 chromosomes (Dujon et al. 1994, Feldmann et al. 1994, Johnston et al. 1994, Bussey et al. 1995, Murakami et al. 1995, Galibert et al. 1996, Bowman et al. 1997, Churcher et al. 1997, Dietrich et al. 1997, Jacq et al. 1997, Philippsen et al. 1997, Tettelin et al. 1997), many of these were published, together with a bioinformatics overview (Mewes et al. 1997), in a special issue of Nature entitled The Yeast Genome Directory (for which the European Commission paid handsomely) that did not appear until the following year. The regular issue of Nature that appeared at the same time carried a ‘News & Views’ piece on the yeast genome from Craig Venter and his colleagues (Clayton et al. 1997) that, in the short run, got more attention than any of those in the Directory. I think that it is difficult for today’s researchers, used to the facility of next-generation sequencing and automated methods of genome assembly, to appreciate just how much work was put in by the different yeast labs involved in the project. When the project began, the sequencing was done by hand (only 7% of the chromosome III sequence was obtained using automated methods (Oliver et al. 1992) and sequence assembly was largely a hand-crafted exercise. Although all yeast researchers, and very many others, make direct or indirect use of the yeast genome sequence every day, it has received the ultimate accolade of ‘citation eclipse’, being rarely referred to in their papers.