Mouse genomic variation and its effect on phenotypes and gene regulation

Abstract

We report genome sequences of 17 inbred strains of laboratory mice and identify almost ten times more variants than previously known. We use these genomes to explore the phylogenetic history of the laboratory mouse and to examine the functional consequences of allele-specific variation on transcript abundance, revealing that at least 12% of transcripts show a significant tissue-specific expression bias. By identifying candidate functional variants at 718 quantitative trait loci we show that the molecular nature of functional variants and their position relative to genes vary according to the effect size of the locus. These sequences provide a starting point for a new era in the functional analysis of a key model organism.

Figure 1. An overview of variants called from 17 mouse genomes relative to the reference

a, The four wild strains (CAST/EiJ, WSB/EiJ, PWK/PhJ and SPRET/EiJ) are representative of the Mus musculus castaneus, Mus musculus musculus, Mus musculus domesticus and Mus spretus taxa and include the progenitors from which the classical laboratory strains were derived. These genomes are shown in a circle with tracks indicating the relative density of SNPs, structural variants (SVs) and uncallable regions (binned into 10-Mb regions). Transposable element (TE) insertions, which are a subset of the structural variant calls, are shown as a separate track. Corresponding tracks are shown for each of the 13 classical laboratory strains to the right of the circle. Links crossing the circle indicate regions on the reference where the wild strain is closest to the reference (375-kb bins). b, The numbers inside the Venn diagrams indicate the number of SNPs, indels, structural variant deletions and transposable element insertions in the wild and classical laboratory strains. The numbers beneath each Venn diagram indicate totals for each type of variant in the wild and classical laboratory strains.

Figure 2. Genomic partitioning of phylogenetic history

Bayesian concordance factors were estimated from 43,255 individual locus trees. 87.9% of the genes place M. spretus (Spret) and rat as the outgroup to the M. musculus subspecies. Within the M. musculus subspecies, there is a primary history supporting a M. m. musculus (Musc)/M. m. castaneus (Cast) sister relationship (37.9%) with M. m. domesticus (Dom) branching off first. The two alternative topologies are supported by equal percentages of the genome (30.3% and 30.2%). 95% credibility intervals on all estimates are ± 0.001.

Figure 3. Allele-specific biases in RNA expression levels between tissues from C57BL/6J x DBA/2J F1 mice

RNA was sequenced from six tissues: hippocampus (hippo), spleen, liver, heart, lung and thymus. Each point represents a gene, and the bias ranges from 1.0 (exclusively C57BL/6J) to 0.0 (exclusively DBA/2J). The tissue comparison is shown above each plot. The points are coloured by whether the difference in bias is not significant (blue), significantly different bias but in the same direction (pink) or significantly different but switching direction (green).

Figure 4. Enrichment of functional variants

Each line shows the ratio of the percentage of functional variants at a QTL over the percentage of variants expected. Ratios greater than one indicate that functional variants are enriched in a classification and ratios less than one indicate a dearth of functional variants. Functional variants are classified by their position relative to a gene and by their molecular class: SNPs, structural variants and insertion/deletions (indels) polymorphisms.