Historically, investigations into how genes influence disease phenotypes have relied on modifying the genomes of inbred, genetically identical mice. In these conventional tests, a scientist tweaks the same single gene in rodents in an experiment to assess how that change affects them. But biologists are increasingly undertaking experiments with genetically diverse mice; Lusis' panel of model animals has four million common single nucleotide polymorphisms of variation, and other models have even more. By using advanced statistical tools to compare the varied genomes of the mice—and simultaneously deploying ever-improving technology to detect the quantity and kinds of proteins in these animals' bodies—they can build a more sophisticated model of how multiple genetic variants lead to disease phenotypes. “It's sort of like connecting the dots,” Lusis says. “It's putting together interactions and building up the architecture of these complex traits.”
In June, researchers from the Jackson Laboratory and Harvard Medical School published a paper in which they used this approach to link different genetic variants to the relative abundance of thousands of different proteins present in mouse livers. The scientists began by quantifying the amounts of mRNA transcripts and proteins in the livers of 192 genetically unique mice bred from eight laboratory mouse lines with fully sequenced genomes (Nature 534, 500–505, 2016). The genetic diversity of the mice resulted in varying amounts of particular mRNA transcripts and proteins in their livers. The scientists analyzed these discrepancies to figure out which genetic variations were correlated with changes in mRNA transcript or protein levels.
This is a preview of subscription content, access via your institution