The Collaborative Cross (CC) recombinant inbred panel was conceived as an ideal resource for mammalian system genetics. The pre-Collaborative Cross (pre-CC) is a proof-of-concept experiment involving CC lines that have undergone at least five generations of inbreeding. Pre-CC mice are partially inbred strains created by intercrossing eight founder (parental) strains: 129S1/SvImJ, (129S1), A/J (AJ), C57BL/6J (B6), CAST/Ei (CAST), NOD/LtJ (NOD), NZO/H1LtJ (NZO), PWK/Ph (PWK), and WSB/Ei (WSB). Sister-brother mating in 220 families was done for 4-11 generations. Siblings from these lines were each involved in one of four distinct phenotyping arms, then genotyped on a high-density Affymetrix platform. The genetic profile of these emerging lines reveals high diversity, balanced allele frequencies, and well-distributed recombination – all ideal qualities for a mapping panel. We have mapped white spot, a discrete trait; body weight, a highly polygenic complex trait; and more than 11,000 liver gene expression traits. These analyses provide a glimpse of the potential mapping power and resolution of the CC. The goal of this project was to identify eQTLs for genome-wide gene expression from liver tissue in pre-CC mice.
Liver gene expression
Gene expression profiling from liver tissue from pre-CC males. One animal was sampled from 157 pre-CC strain. Hybridizations were performed at the Gene Expression Core Facility at The Jackson Laboratory. Samples were divided in six batches of about 11 to 24 samples. Two batches were performed per day. Probe-level data was summarized by a custom CDF file based on Ensembl genes (NCBI 37, Ensembl 49; htttp://brainarray.mbni.med.umich.edu). The parental strains are not part of this submission.
Upon completion of the pre-CC experiment at 12–16 wk of age, the presence of a white head spot was noted. coatcolor_phenotypes.txt
Pre-CC mice were weaned at Oak Ridge National Lab (ORNL) and transported to University of North Carolina at Chapel Hill (UNC) at 9–13 wk of age. Baseline body weight was measured after acclimating at UNC for 1 wk. bw_pheno.txt
Each mouse in the pre-CC experiment was genotyped using a highdensity SNP array. Most of the genotyping was completed using ‘‘test’’ arrays. These arrays were developed as an intermediate step in the process of developing the Mouse Diversity array (Yang et al. 2009). There are two versions of the test array: A-array and B-array. The A-array includes 294,878 SNP assays, and the B-array contains 287,687 additional SNP assays. We determined that 181,752 (A-array) and 180,976 (B-array) SNP assays performed well and targeted loci that are polymorphic among the eight founder strains. There is no overlap between the two arrays, but the genome coverage is complete and uniformly distributed in both. In some cases, animals from the same phenotyping arm were genotyped with different arrays. Integration was achieved by merging the two sets and using an HMM to impute haplotypes at loci with missing genotypes. Due to the high marker density, this procedure was very effective. The exercise behavior and metabolism arm was completely genotyped with the A-array. Genotype data is publically available for download at the CC Status website (http://csbio.unc.edu/CCstatus/index.py).
Note: Within each csv file you will observe a header for each column of genotypes. The first few link the corresponding mouse strain to a letter - such as the letter A to the strain A/J - and so forth, until the letter H. The following columns correspond to crosses. Their headers are given as sequences of letters, such as ECBDAHFG, ADFEHGBC, and BCGDFEAH. The gender order within these sequences is FMFMFMFM where F = female and M = male. The Y chromosome comes from the last position, mitochondria from the first. Array genotyping data: metabolism filtered genotypes 4hmm.zip