What lab mice are made of

Researchers have created genome-wide, high-resolution genetic maps of 100 inbred mouse strains. Compared with wild-derived strains, most of today's classical lab strains have limited genetic diversity and overwhelmingly originate from a single species of domestic mouse, the researchers report.

Fancy mice, Mus musculus domesticus Wikimedia Commons

The findings resolve competing views on the origins and composition of laboratory mice. The paper, published online this week in Nature Genetics, also unveils an online tool with which scientists can visualize detailed genetic data and phylogeny trees of inbred and wild-derived mouse strains to decide which strains are likely to be most useful in an experiment.

"It's a really nice paper. It sorts out the controversy of what the background of these animals really is," said Elizabeth Fisher, a neuroscientist at University College London who was not involved in the research. As for the online tool, she adds, "I'm sure we'll all be using it."

"These are mice that people have been working with for more or less 100 years," said senior author Gary Churchill, a mouse geneticist at the Jackson Laboratory in Bar Harbor, Maine. "It's embarrassing to have such a highly used model organism and not really understand what its relationship is to the wild."

In 2007, two research groups came to different conclusions about the origins of classical inbred mouse strains. One group found that the strains are composed of 68 percent Mus musculus domesticus -- "fancy" domesticated mice derived from Western European mouse breeders -- 19 percent other known strains, and 13 percent of unknown origin. Using the same public data, a second group, led by Churchill, concluded that inbred strains are far less diverse: 92 percent of their genetic makeup is derived from M. m. domesticus, with the remaining 7 percent originating from two other known species.

Over the next three years, Churchill, with collaborator Fernando Pardo-Manuel de Villena at the University of North Carolina at Chapel Hill, designed a high-density genotyping array and software to genotype hundreds of mouse strains at high resolution. Once the technology was ready, the team collected 36 wild-derived strains of mice -- "There were people literally going out in the bushes and trapping mice," said Churchill -- and analyzed their genetic data to determine which species carried which single nucleotide polymorphisms (SNPs) -- single-letter variations in the DNA. These data were then used to identify the origins of 100 strains of classical inbred lab mice.

They found that the genomes of inbred lab strains are overwhelmingly M. m. domesticus in origin -- 90 percent or more in every case -- with the remaining percentage almost exclusively due to a Japanese species of mice. "We recapitulated what we said in 2007, so it was nice to be vindicated," said Churchill. And inbred lab strains had significantly less genetic diversity compared to wild-derived strains: standard lab strains carry about 12 million SNPs, while the offspring of inbred mice and wild-derived strains (part of a 2004 project launched by Churchill and Pardo-Manuel de Villena called the Collaborative Cross) have 45 million SNPs.

There are even regions of the lab mice genomes that are genetically "blind," said Churchill, where there is no genetic variation at all. Without variation, there is no way for researchers to test the effects of different variants. For example, chunks of chromosome 10, where genes implicated in lifespan have been identified, are identical in lab strains. Such regions may encourage researchers to incorporate more wild-derived strains into their work, for allelic diversity at those locations, said Churchill. "You can never have too much diversity."

In addition, the team compiled data from 162 mouse strains to create a publically available online tool called the Mouse Phylogeny Viewer. On the site, researchers can pick any set of strains, compare how they are related, view their origins, and even look at specific regions of the mouse genome to see how strains compare at individual alleles.

The new data and online tool should help researchers incorporate new and more diverse mouse strains into their research, said Fisher. With the new tool, wild-derived mice "are not such a scary unknown," she said. "That's a wonderful thing to come out of this paper, that there will probably be an uptake of more diverse strains."

H. Yang, et al., "Subspecific origin and haplotype diversity in the laboratory mouse," Nat Genet, doi:10.1038/ng.84, 2011.


Related stories:

Model organisms up close
[22nd December 2010]

Of men, not mice
[28th May 2009]

Lab mouse genome isn't simple
[22nd June 2004]