When the first complete genome of a living organism, the bacterium Haemophilus influenzae, was sequenced in July 1995, there were mixed reactions from the biological research community  [1]. Some welcomed it as a major event which opened up radically new avenues in the study of life, while others saw it as at best a purely technological and economic exploit, which distracted the attention of decision-makers and the public away from the real concerns of research. Five years later, now that the genome sequences of several dozen bacteria and three eukaryotic organisms - the yeast Saccharomyces cerevisiae, the nematode Caenorhabditis elegans and the fruit-fly Drosophila melanogaster - have been obtained and published, and a draft version of the human genome sequence has been announced, there are still the same differences of opinion  [2] [3].

The strategy employed is both systematic and exploratory, and it is this two-pronged approach which has prompted the debate. The expression "post-genome", misleading in more than one respect, is often used to mark the end of a period of blind experimentation and to welcome the return to a hypothetico-deductive type of approach. It gives the impression that since the beginning of the 1970s, sequences have simply been collected, without any information about the function and evolution of living systems being learnt from them. In fact, the availability of sequences merely marks the beginning of the long and difficult job of analysing the data, frequently interrupted by a return to experimentation and even to new sequencing. How should this mass of information be used, to turn it into biological knowledge? A sequence has a formal structure and lends itself naturally to analysis by computer. What roles should computational analysis of genomic data and experimental approaches play?