부산대학교 도서관

Summary: Breast cancer is caused by the accumulation of genetic mutations, ranging from single base-pair alterations to large-scale chromosome rearrangements. Research platforms must be able to detect and characterize this wide range of mutations for not only the comprehensive identification of genes that contribute to tumorigenesis, but also tumor characterization in the clinical setting. Here we present optical mapping as a system for high-resolution structural characterization of breast cancer genomes. Optical mapping creates ordered restriction maps from single DNA molecules which are then assembled to create a genome-wide physical map with a resolution of ∼20 kb. We analyzed the model breast cancer genomes MCF7 and HCC 1937 using optical mapping, which combined contain over 1,000,000 single molecule maps. We were able to identify 2,441 discrete structural alterations, ranging from single basepair changes to genomic insertion and deletion up to 200 kb. Moreover, we were able to identify 59 large-scale rearrangements, including chromosome translocations. These results have been validated using direct experimental as well as comparative analysis. From this diverse collection of structural mutations, we have identified 86 candidate breast cancer genes, RUNX1---the most frequently rearranged gene in leukemia. We have also identified 7 candidate fusion genes without the need for addition downstream experimentation. Because optical mapping does not rely on clone libraries and sequencing regimes, it will be a useful platform for structural characterization and gene discovery in all types of cancer genomes.