The End of the Beginning: The Race to Begin Human Genome Sequencing

The Human Genome Project has different meanings to different scientists, but the essence of it has always been the drive to obtain the total reference genome sequence for humans. In the first five years of its official existence, the Human Genome Project has concentrated on the development of genetic and physical maps that would ultimately support genome sequencing in the human and a few model organisms. The comprehensive genetic and physical maps, and the associated reagents, have already found use amongst a large, diverse, and enthusiastic group of geneticists who are using them to find genes for both simple and complex hereditary traits. One revolution is already under way: Geneticists are increasingly turning their attention to the dissection of complex traits per se using maps, rather than using the classical approach of analyzing only Mendelian traits associated with complex phenotypes. It is now time to set attention to the main task, that of large-scale human genome sequencing, and that is precisely what has been rumbling in the background. In the past year there has been a clear transition to the thought that current technologies are adequate to begin the task of human genome sequencing, since radically new and revolutionary sequencing technologies are not evident and there is greater danger in not having the sequence sooner. It is now increasingly clear that a further revolution in biology will arise from the sequence itself, not necessarily from the technologies used to obtain it. Consistent with this view, some countries have already initiated largescale human genome sequencing programs. Without much fanfare, the race to finish the human sequence has begun. The transition to fulfilling the sequencing goals has been largely prompted by a proposal made by Robert Waterston at Washington University in the United States and John Sulston at the Sanger Centre in the United Kingdom, based on their experience in genome sequencing of Caenorhabditis elegans. Agencies funding the Human Genome Project seem to agree as much. In the United States, the National Institutes of Health (NIH) has funded six groups to establish pilot programs and develop production groups capable of sequencing large tracts of human DNA at high accuracy and low cost. In the United Kingdom, the Sanger Centre has been funded by the Wellcome Trust to support the production of first-generation sequence for up to one-third of the human genome. And there are evolving sequencing efforts in Germany and France. Re-establishing human genome sequencing as the current priority has led to several discusions in the sequencing community on specific issues that require immediate attention: these relate to sequencing strategy, data quality, and data release. Over the last six months, three meetings have examined these issues. In February, the Wellcome Trust sponsored a meeting in Bermuda, attended by the major practitioners of large-scale sequencing, with the objective of establishing policies for data release and implementation of a system for coordinating sequencing targets. A general policy for coordinating efforts to prevent unnecessary duplicat ion of sequencing was endorsed, but, important ly, there was widespread consensus on the desirability of "immediate and free" data release. A second meeting, at the NIH in March, emphasized that a premium should be placed on generating high-quality data. The ways in which such a standard would be maintained were less clear, nor were practical mechanisms for monitoring data quality established. A third gathering, at the annual Cold Spring Harbor Genome Mapping and Sequencing meeting in May, clarified the increasing convergence on strategies for performing large-scale genome sequencing. In fact, the differences in strategies are all subtle since virtually all major sequencing groups use fluorescentbased four-color chemistries and some variation of methods that employ both random shotgun and directed phases. These meetings have led to a consensus view that the general strategies for sequencing are largely established, even though numerous technical details need to be resolved. With this in mind, a few groups have made strong commitments to sequence specific regions of the genome and a number of consortia have been formed. For some groups the path forward is clear, with little left to decide; the major impediment is obtaining funds to operate large-scale sequencing programs. These scientists argue that from this activity itself will emerge not only the required experience but also all of the required information