Panos Deloukas: mapping SNP haplotypes. Credit: WELLCOME PHOTO LIBRARY

As the Human Genome Project winds down, many researchers are stepping over into the next international public endeavour, the haplotype map project, or HapMap. This US$100-million initiative was launched in October 2002 by the US National Human Genome Research Institute to provide the tools that will speed up the discovery of genetic factors that contribute to common conditions such as diabetes and heart disease. The results will be made quickly and freely available on the Internet.

Along the genome, certain combinations of alleles at nearby SNP sites have remained unchanged over many generations and are inherited together as halotypes. The aim of HapMap is to build a genome map of these haplotype blocks from different ethnic groups, mainly from China, Japan, Nigeria and northern Europe.

Labs in Canada, China, Japan, the United Kingdom and the United States will be involved. “The goal of the project is to validate SNPs across the entire genome and build tools to enable association studies, to compare genetic markers in people who have a disease with markers in a control population,” says Panos Deloukas, senior group leader at the human genetics department at the Wellcome Trust Sanger Institute near Cambridge, UK. The HapMap, set up as an international public–private research consortium, is expected to take three years to complete.

Why bother with a HapMap when whole-genome scanning technologies already exist? “With the technologies we have now, scanning the whole genome gene by gene is very costly. Scanning the genome haplotype by haplotype would be 5% of the cost,” points out Julio Licinio at the pharmacogenomics lab of the University of California, Los Angeles.

An alternative way of reducing the costs of whole-genome scanning has just been launched by the discovery-genetics company Sequenom in San Diego, California. Its Allelotyping technology uses high-resolution MALDI–TOF (matrix-assisted laser desorption/ ionization–time-of-flight) mass spectrometry to analyse the frequency of a SNP allele in large DNA pools of up to 500 individual samples. Only 50 picograms of DNA from each individual are required, and allelic frequencies as low as 3% can be detected. “Our customers can afford to do whole-genome scans today, because whole-genome scans are affordable on pools but not on individuals,” says Charles Cantor, chief scientific officer at Sequenom.