The Human HapMap and Personalized Medicine

In the October 27th issue of Nature, an international team of scientists published a high-density haplotype map of the human genome. It marks the conclusion of the first phase of the International HapMap Project - a five-year sequel to the Human Genome Project that began in 2001. The project's goal was to identify the patterns of human genetic variation using single-nucleotide polymorphisms (SNPs). Although the DNA sequence humans is 99.9% identical between individuals, there are more than 10 million SNPs in the world's population. Because SNPs that are near one another tend to be passed along from parent to child, a correlation arises between them. By studying that correlation across the genome, HapMap scientists have constructed a map of the underlying structure of genetic variation.

Applications of the HapMap

The human haplotype map is an important tool for studying complex phenotypes like disease risk and pharmaceutical response. Before the HapMap Project, choosing the best SNPs for an association study was an extraordinarily difficult problem. There was no way to know which would prove to be polymorphic in the sample population, and which SNPs could capture the most information for a genomic region. Now, investigators can use the patterns in variation to select a small number of markers that represent large genomic regions. This reduces the cost of performing association studies while making them more efficient.

HapMap Genotyping at Washington University

Part of the Phase I data was generated at Washington University School of Medicine in St. Louis. The SNP Research Facility of Dr. Ray Miller was one of six genotyping centers that participated in Phase I of the HapMap. In collaboration with Dr. Pui Kwok's group at UCSF, Miller's lab generated over 1 million genotypes on the short arm of human chromosome 7p.

The Outlook for Personalized Medicine

Researchers field of pharmacogenetics investigate how an individual's genotype affects his or her response to medication. This is especially desirable for drugs like warfarin, a widely prescribed anticoagulant (anti-clotting) medication. The correct warfarin dose varies widely from person to person, and too high of a dose can cause bleeding in the patient. Dr. Brian Gage of Washington University Medical Center, a pioneer in warfarin pharmacogenetics, has developed a web site (www.warfarindosing.org) to help physicians estimate the therapeutic dose in patients new to warfarin. Estimates are based on clinical factors as well as the genotypes of two genes that are known to affect warfarin dosing. This offers an excellent example of how the most accurate phenotype predictions take both environmental and genetic factors into account. As our understanding of the genotypes that underlie drug response grows, studies like these will bridge the gap into a new era of personalized medicine.