Kevin MacDonald, Ph.D.
CSULB, Department of Psychology
Genome Variables
Study Breaks New Ground on Variations in Genome
By NICHOLAS WADE
New York Times
July 13, 2001A large-scale study of the variability in the human genome has shown that each human gene may come in 12 different versions on average. The authors also say their findings cast doubt on the way that a large government and industry program is mining the genome for the genetic basis of common human diseases.
The study was undertaken by Genaissance Pharmaceuticals, a biotechnology company in New Haven, to discover the genetic basis for why individuals respond differently to given drugs.
The company's plan is to help doctors determine through a genetic test, rather than trial and error, which asthma or cholesterol drug, for example, would be best for a patient. The company has bought 60 of the latest-model DNA-sequencing machines, one of the largest sets outside those of the genome-sequencing centers, and installed them in a refurbished World War II munitions plant in New Haven.
The decoding of the human genome provided, in effect, a single DNA sequence. But what is of interest for most medical purposes is to know how each gene varies from one person to another and how those variations influence an individual's susceptibility to disease and response to drugs.
Government and industry have devoted considerable resources to looking for sites in the genome's three- billion-unit sequence where some people have one letter of DNA and some another. These sites of common variation are known as SNP's (pronounced snips). The hope is to correlate SNP's with disease and discover which versions of a gene predispose a person to diseases like diabetes or cancer.
So far three million or so SNP's have been cataloged, but they have been discovered more or less at random across the genome. Genaissance officials estimate that there are some 30 million SNP's in the human population, and say a SNP chosen at random is unlikely to pinpoint a gene variant that causes disease or drug response.
Instead, Genaissance is analyzing sets of closely bunched SNP's in the hope of correlating the patterns with a patient's response to various drugs. A connected set of SNP's inherited as a unit is called a haplotype, and so the company calls its method the haplotype approach.
Its ambition is to catalog the haplotypes of every human gene by decoding each gene in a total of about 90 people drawn from the three major human-population groups: Africans, Asians and Europeans. Company officials say they have already sequenced 4,000 of the estimated 30,000 human genes in this way and intend to sequence all of them within a few years.
Genaissance has published the results from 313 of these genes in today's issue of Science, in an article by Dr. J. Claiborne Stephens and several colleagues. From an analysis of the SNP's in these 313 genes, the authors calculate that each gene exists in 12 versions on average.
Dr. Kenneth Kidd, a population geneticist at Yale University who was not connected with the study, described the data as very important and said it confirmed how much genetic variability existed in the human population. He said he agreed with Genaissance's view that the SNP approach was "misconceived," and chided the government for having stripped ethnic identities from the panel of people whose genomes have been searched for SNP's.
But Dr. Francis Collins, director of the National Human Genome Research Institute, said the SNP and haplotype approaches were not in opposition. His institute plans a workshop next week, he said, on how to build a haplotype map of the genome. There may be quick ways of arranging the SNP's already discovered into the sets or haplotypes in which they are usually inherited.
Dr. Gualberto Ruaˆ±o, chief executive of Genaissance, said its haplotype catalog could help pharmaceutical companies to profile patients who respond best to their drugs, and to develop new drugs for any identified group of those who fail to respond.
Genetic information can be a delicate matter, particularly if it reveals a person's susceptibility to disease. Dr. Gerald F. Vovis, the company's chief technology officer, said Genaissance was developing information only about drug response, which in his view would fall into a less delicate category.
Both Dr. Kidd and Dr. David Altshuler of Harvard University said Genaissance was overcounting the number of SNP's and haplotypes, by including SNP's found in a single person. Because only common SNP's are likely to play a role in common diseases, geneticists usually count a DNA change as a SNP only it if occurs in at least 1 percent of a population. Many of the DNA changes Genaissance finds are singletons of perhaps no significance for any but the individual involved.
