The laboratory mouse Mus musculus, the most commonly used model organism for pre-clinical studies, has recently become even more valuable for research in human diseases. A study appearing in the July 2007 issue of Nature published the first mouse haplotype map that catalogs nearly 8.27 million single nucleotide polymorphisms (SNPs) in 15 commonly used strains of laboratory mice. The data is available on the National Center for Biotechnology Information Web site at http://www.ncbi.nlm.nih.gov/SNP/. Perlegen Sciences, the Mountain View, California company that performed the study, also makes the information available at http://mouse.perlegen.com. The sites allow researchers to download SNPs, genotypes, and long range-polymerase chain reaction (LR-PCR) primer pairs, which are currently mapped to NCBI Build 36.

The project, which took about 2 years to complete, was funded by the National Institute of Environmental Health Sciences (NIEHS), part of the National Institutes of Health (NIH), to better understand how differences in genetic make-up can contribute to differences in susceptibility to certain disease-causing environmental agents. Since 99% of mouse genes have a human counterpart, this data can hopefully provide better insight into why one human is prone to a certain disease, while another in the same environment remains disease-free.

“Genetic variability in the response to toxic chemicals is one of the big problems that the National Toxicology Program (NTP) faces,” says Frank Johnson, Ph.D. research geneticist in the Toxicology Operations branch at the NIEHS in Research Triangle Park, NC. “We are going to look at some of these same mouse strains in toxicological experiments, look for genetic variation and select strains at either end of the response range and study the genotype-phenotype associations.”

The NTP had reached out to various technology vendors and service providers for help with this study and the contract was eventually awarded to Perlegen. “We could do it in a cost efficient manner because we took advantage of a high-density oligonucleotide array technology developed by Affymetrix,” says Kelly Frazer, Ph.D., who was then the Vice President of Genomics Biology at Perlegen Sciences and one of the lead investigators on the project. Frazer has since left Perlegen and is now an associate professor in the Department of Molecular Biology and Experimental Medicine and Director of Genomic Medicine at the Scripps Research Institute in La Jolla, CA.

Ancestral map for a region of chromosome 14 for 12 commonly used laboratory strains. Each segment is colored to represent origin for a specific mouse subspecies. As can be seen, in each region of the genome different laboratory strains inherited their DNA from different mouse subspecies. (Source: Eleazar Eskin)Click to enlarge.

The mouse strain C57BL/6J, whose complete DNA sequence is available, was used as a reference to conduct the re-sequencing of four wild-derived and eleven classical mouse strains. “We took the reference sequence for C57Black6J and tiled the entire sequence down using a series of 25 mer nucleotides,” says Frazer. A long range PCR method developed by Perlegen was used to amplify the genomes of the individual strains to be hybridized to the array. The hybridization patterns then revealed the differences in the sequence between the amplified and the reference strain. “Where the patterns differed is where we looked and identified polymorphisms,” says Frazer.

The task at hand was quite daunting. “We were dealing with 15 different samples but each had over 200,000 PCR reactions done on it and they had to be pooled and hybridized,” says Frazer. “So we knew that there would be human errors, and the most important thing was to make sure that you can identify the errors that you know are going to happen.”

The other challenge was handling the sheer volume of data. “With 8 million SNPs and 15 strains it was just a lot of data,” says Eskin Eleazar, Ph.D., assistant professor in the department of Computer Science and Human Genetics at the University of California in Los Angeles, who helped Perlegen with the data analysis.

Eleazar’s group also built the web-server and the ancestral map browser. “We looked at each segment in the classical in-bred strain and tried to find its ancestor,” says Eleazar. “Most of the variation in the current lab strains can be traced back to the ancestral population that contributed to these strains.”

Now that the haplotype patterns and individual SNPs have been identified in these 15 mouse strains, many others can be genotyped using just a fraction of the SNPs, while the rest can be deduced based on the patterns. “What we have generated is a fantastic resource for the mouse genome community to move forward on and generating the resource is only just the beginning,” says Frazer. “There are multiple other resources being generated in the mouse community and the combining of all these resources is going to lead to incredible discoveries.”

Funding Opportunity