Although geneticists would not survive without mutations, they could easily live without the tedious hassle of finding them. And, generally speaking, the bigger the model organism, the harder the slog. But things might be set to change, as Benjamin Kile and colleagues now report the result of a systematic and relatively straightforward way of identifying mutations in mouse genes. By borrowing a technique used in fruitflies, they have identified 88 new point mutations on a portion of chromosome 11 — and all by simply crossing mice and scoring the colour of their coat.

Among model systems, mice are arguably the most promising organisms in which to study the function of human genes, particularly those that are involved in disease. However, traditional methods of identifying loci by screening for a mutant phenotype often involve large pedigrees, low efficiency, laborious mapping tools and difficulty in maintaining the mutant strains. The new approach, published in Nature, seems to have solved these problems. At the heart of it is the use of a so-called 'balancer' chromosome, which, owing to engineered inversions, allows mutations on its homologue to be inherited without recombining.

The authors created a mouse chromosome 11 containing a 24 cM inversion that included ∼700 genes and could be detected phenotypically by the presence of a dominant mutation ( Agouti ) that gives mice yellow fur. The crossing scheme — which is used routinely in flies — goes as follows: a wild-type male that has been mutagenized using the chemical mutagen N-ethyl-N-nitrosourea (ENU) is crossed to a balancer-carrying (yellow) female. Yellow-furred F1 mice, some of which contain a balanced mutation on chromosome 11, are then crossed to a mouse with a balancer plus a chromosome that confers a dominant curly-coat phenotype. The only offspring in the F2 generation to have a yellow non-curly coat are those with a balanced ENU-mutagenized chromosome 11; these can now be mated to each other to characterize the phenotype of the homozygous mutation and to maintain the mutant line.

Of the 88 new mutations that were found in this way, 55 were lethal, whereas the remainder had phenotypes that affected various organs, including the skin, nervous system and blood cells. Balancer chromosomes are now being generated for other regions of the mouse genome, making phenotype-driven screens as easy as they are in flies: mutations are identified in only three generations, are mapped without further crosses and stocks are maintained without the fuss of molecular genotyping. As Janet Rossant points out in a related News and Views piece, such screens will contribute enormously to annotating the mammalian genome — but only if they are accompanied by much needed improvements in phenotyping tools.