New York

Take the strain: a catalogue of genetic differences between mouse breeds will aid disease research. Credit: NIEHS/NIH

Efforts to find genes involved in human disease are set to benefit from a plan to sequence the genetic code of 15 different mouse strains.

Lab mice might look alike, but in fact scientists work with at least 100 different inbred strains. These strains differ hugely in their behaviour, physiology and susceptibility to disease, and researchers are keen to find the differences in the genetic sequences that underlie these traits, with the aim of locating counterparts in humans.

The sequencing project plans to decipher the genomes of 15 mouse strains within two years. It will be largely funded by the US National Institute of Environmental Health Sciences and the sequencing will be done by Perlegen Sciences, based in Mountain View, California. “People are delighted,” says mouse geneticist Kenneth Paigen of the Jackson Laboratory in Bar Harbor, Maine.

Researchers have long puzzled over the differences between mouse strains (see Nature 415, 8–9; 2002). After a few weeks on a high-fat diet, for example, one strain's cholesterol level hovers at a healthy 100 milligrams per decilitre, whereas another's rockets to more than five times that amount.

Tracking down the genes involved in conditions such as high cholesterol, diabetes, obesity or cancer is a long process. Researchers cross a mouse strain susceptible to diabetes, for example, with a disease-resistant strain, and identify the chunks of the genome that are inherited with the disease. They then laboriously scour the DNA in these regions for ‘susceptibility genes’.

The new line-up of mouse genomes should accelerate the process significantly. The genomic data will show researchers exactly how the DNA of the diabetes-prone strain differs from that of other strains. So researchers should be able to home in quickly on relevant genes and perhaps find corresponding ones involved in human disease.

Geneticists have had access to the genome of one widely used mouse strain, called C57BL/6J, since 2002 — and, for a fee, genomes of a few other strains are available from Celera Genomics in Rockville, Maryland. The project's organizers say this is the most comprehensive attempt yet to analyse the genomes of different strains, and that the code will be deposited in public databases.

The 15 strains were selected because they are widely used in research, are evolutionarily diverse and show different propensities to common diseases.

Researchers hope to link their genetic information to the Mouse Phenome Project, an international effort launched in 2000 to collate the anatomical, physiological and behavioural differences between mouse strains. This should help researchers to choose the most appropriate strain for studying a disease.

Ultimately, the amount of detailed biological and genetic data collected might be so great that it eliminates the need for many mouse crosses altogether, says the leader of the Phenome Project, Molly Bogue, also at the Jackson Laboratory. Researchers could simply select a panel of mouse strains with different cholesterol levels, and use computer analyses to help pinpoint the candidate genes that differ between them. “That's what we're hoping for,” she says.