Map-based cloning is an iterative approach that identifies the underlying genetic cause of a mutant phenotype. However, the classic protocol of positional cloning is time-consuming and labor-intensive. We now describe a genome sequence-based cloning approach that has led to localizing the underlying genetic cause of spontaneous fractures (sfx) in a mouse model. The sfx/sfx mouse is characterized by a spontaneous femoral fracture seen around 6 weeks of age, which represents a new mouse model for bone fragility. Genetic studies indicate that the phenotype of sfx/sfx mice is caused by an alteration at a single locus that is roughly mapped onto the central region of mouse Chromosome 14. Using our strategy of combining mouse genome resources and high-throughput technology, we discovered a deletion of all 12 exons in the gene for L-gulonolactone oxidase (LGO), a key enzyme in the synthesis of ascorbic acid. We have also examined the expression of LGO and found no expression of LGO in sfx mice while the LGO expresses in several tissues of normal mice. Our data demonstrated the feasibility to positionally clone the mutated gene from a non-fine-mapped locus, which has applicability to the positional cloning of genes from many other animal models, as their genome sequences are sequenced or will be sequenced soon.