The genetic structure of recombinant inbred mice: high-resolution consensus maps for complex trait analysis

Robert W Williams¹^,2^,3 , Jing Gu¹^,3, Shuhua Qi²^,3 and Lu Lu¹^,2^,3

¹Center of Genomics and Bioinformatics University of Tennessee Health Science Center, College of Medicine, Memphis, TN 38163, USA

²Center for Neuroscience University of Tennessee Health Science Center, College of Medicine, Memphis, TN 38163, USA

³Department of Anatomy and Neurobiology, University of Tennessee Health Science Center, College of Medicine, Memphis, TN 38163, USA

author email corresponding author email

Genome Biology 2001, 2:research0046.1-0046.18doi:10.1186/gb-2001-2-11-research0046


Published:	22 October 2001

A previous version of this manuscript was made available before peer review at http://genomebiology.com/2001/2/8/preprint/0007/

Subject areas: Genetics, Methods, Model organisms

Abstract

Background

Recombinant inbred (RI) strains of mice are an important resource used to map and analyze complex traits. They have proved particularly effective in multidisciplinary genetic studies. Widespread use of RI strains has been hampered by their modest numbers and by the difficulty of combining results derived from different RI sets.

Results

We have increased the density of typed microsatellite markers two- to five-fold in each of several major RI sets that share C57BL/6 as a parental strain (AXB, BXA, BXD, BXH and CXB). A common set of 490 markers was genotyped in just over 100 RI strains. Genotypes of around 1,100 additional microsatellites in one or more RI sets were generated, collected and checked for errors. Consensus RI maps that integrate genotypes of approximately 1,600 microsatellite loci were assembled. The genomes of individual strains typically incorporate 45-55 recombination breakpoints. The collected RI set - termed the BXN set - contains approximately 5,000 breakpoints. The distribution of recombinations approximates a Poisson distribution and distances between breakpoints average about 0.5 centimorgans (cM). Locations of most breakpoints have been defined with a precision of < 2 cM. Genotypes deviate from Hardy-Weinberg equilibrium in only a small number of intervals.

Conclusions

Consensus maps derived from RI strains conform almost exactly to theoretical expectation and are close to the length predicted by the Haldane-Waddington equation (x3.6 for a 2-3 cM interval between markers). Non-syntenic associations between different chromosomes introduce predictable distortions in quantitative trait locus (QTL) datasets that can be partly corrected using two-locus correlation matrices.