News Notes

HGP Enters Final Phase: Even as the Human Genome Project (HGP) enters its final phase, which officially began May 9, perhaps the most fundamental bit of genomic trivia persists: Exactly how many human genes are there? At the 13th annual Cold Spring Harbor Meeting on Genome Sequencing and Biology held May 10, scientists marveled at the astounding acceleration of sequencing efforts in the last year. Participants, including all 16 leaders of the HGP sequencing centers, enjoyed the most recent sequencing milestone: the completed DNA sequence of chromosome 21, the smallest human chromosome (M. Hattori et al., "The DNA sequence of human chromosome 21," Nature, 405:311-9, May 18, 2000). They also, according to Robert Waterston, director of the genome-sequencing center at Washington University in St. Louis, lightheartedly took bets on exactly how many genes actually make up the human genome. Estimates have been as high as 150,000, but based on extrapolations from the averaged gene density of chromosome 21 (which has a paltry 225 genes) and that of chromosome 22 (which has about 545 genes) reported last December, the number would be closer to 40,000. The lower numbers raise the intriguing possibility that humans have as few as four times as many genes as fruit flies have and eight times as many as yeast has. Investigators won't know the actual number for sure until the project is finished in 2003; the "working draft," which could cover as much as 90 percent of the euchromatic portion of the genome, is ongoing, slated for completion at the end of this June. The final phase, though, won't just provide a gene count, but, says Waterston, will reduce uncertainty in all of biology by providing a standard, a "solid foundation for biology." Says Richard Gibbs, director of the Baylor College of Medicine Sequencing Center, "The real utility of the data doesn't come into its own until the product's finished," adding that even a little uncertainty can cause problems when trying to predict genes, do automatic comparison with other sequences, or find genetic variation. Companies Mark Their Territory: Rockville, Md.-based Celera Genomics Group announced at a congressional hearing in early April that it had finished sequencing pieces of the human genome and planned to assemble those pieces by the end of June. Assuming that it can efficiently incorporate the publicly available HGP data, a feat it has only demonstrated in the much less complex fruit fly, Celera's map will have better coverage than the HGP's "draft," though it will still not be complete. Waterston points out that although Celera is on record as saying it will release data quarterly, it hasn't yet done so, something the public effort continues to frown upon. But Gibbs suggests that the government vs. Celera hype is dying down. "The whole relationship's getting clearer and clearer," he says. Meanwhile, companies such as the Oakland, Calif.-based Doubletwist have already done a good deal of "data mining" based on the available HGP sequence. On May 8, it announced that it had completed an extensive annotation, or computational analysis, of the data available so far, and the company estimates that the human genome has about 100,000 genes. Its results will be available to scientists on a subscription basis. Waterston cautions, though, that annotating the genome, much like translating a foreign language, is not an exact science: "Annotation is a beginning art. It's important that there be some public standards out there." Meanwhile, at the SNP Consortium: Last year the SNP Consortium brought together pharmaceutical industry giants, academic centers, and the nonprofit Wellcome Trust in an ambitious two-year endeavor: to create a database of 300,000 single nucleotide polymorphisms, or SNPs, which are points along a person's DNA map that vary from one individual to another (E. Russo, P. Smaglik, "Single nucleotide polymorphisms: Big Pharma hedges its bets," The Scientist, 13[15]:1, July 19, 1999). The goal was not to hoard this information or sell it, but to make it publicly available. "Our first year has been very successful. We have publicly released [more than] 102,000 mapped SNPs, and another 50,000 will be mapped in the near future," comments Arthur Holden, chairman and CEO of the consortium. The consortium, taking advantage of the advanced HGP, is randomly sequencing across the genome, a method called whole genome shotgun. The sequences can then be aligned against the publicly available human genome sequence to identify SNPs. The plan is to identify a half-million to a million SNPs by the end of this year and map 90 percent of them by the end of the second year. Holden indicates this will increase the chances of finding markers that will lend themselves to effective genotyping or disease association. Stephen Liggett, professor of medicine at the University of Cincinnati, is one of many researchers using SNPs to study disease and individual drug response. He suggests checking all the available SNP databases before undergoing a SNP discovery program. "But you need to know where that data came from, including how many people were actually genotyped and their racial composition," he warns. For example, some SNPs are highly ethnic-group dependent. A database developed from Caucasians will miss some low-frequency SNPs that are found in Africans or Asians. "My own feeling is the SNP consortium is a little shy on ethnic populations. From a resource standpoint, they had to draw the line someplace," he adds. The consortium database is derived from the genomes of 24 ethnically diverse individuals, including Caucasian, African American, and Asian. "That diversity has always been fundamentally part of our strategy, and it is one of the major limitations of other SNP databases," responds Holden. Liggett points out that the SNP Consortium is looking at random SNPs through the genome. The likelihood that a SNP from the database will be in the coding or promoter region of a gene is low. These SNPs are probably directly related to disease because they affect the proteins produced by the body. "If you are searching for a specific series of SNPs in a gene, you can hope by luck it will be in the database," comments Holden. Alternatively, a SNP might not be in the coding region of a gene, but it may be close to a SNP that is associated with disease. With a large-enough clinical study, statistically there will be a linkage between the disease trait of interest and the SNP even though the SNP is not responsible for the disease. Liggett feels that databases from companies doing targeted SNP searches may be more useful than databases made up of random SNPs. Incyte Genomics Inc. of Palo Alto, Calif., CuraGen Corp. of New Haven, Conn., and Genaissance Pharmaceuticals Inc., also of New Haven, are such companies. According to Mark Vincent, director of corporate communications at CuraGen, its database of 120,000 SNPs is derived from cDNAs, and SNPs are primarily from the coding region of genes. Many academic labs are also doing targeted SNP searches. Researchers such as Liggett are using the massive amount of biology information known about diseases to choose and survey candidate genes completely for SNPs. The bottom line is translating SNPs to the clinical care of patients, but that is a long way off. "We don't know enough about variants and linkage association of markers with common disease. There is a lot of work to be done. This is a fundamentally new platform that we will all [use] for many, many decades," comments Holden. --Eugene Russo and Nadia S. Halim

News Notes

Become a Member of

Meet the Author

Eugene Russo

Published In

May 2000

Related Research Resources

Bunsen Burners and Bad Hair Days

Research Resources

Podcasts

Webinars

Videos

Infographics

eBooks

Taking the Guesswork Out of Quality Control Standards

PFAS: The Forever Chemicals

Unlocking the Unattainable in Gene Construction

Comparing Analytical Solutions for High-Throughput Drug Discovery

Products

Product News

Thermo Scientific™ Centrifuges with GreenCool Technology

Singleron Biotechnologies and Hamilton Bonaduz AG Announce the Launch of Tensor to Advance Single Cell Sequencing Automation

Zymo Research Launches Research Grant to Empower Mapping the RNome

Cytosurge Appoints Magid Haddouchi as Chief Commercial Officer