Using large-scale molecular data sets to improve breast cancer treatment.

The treatment of breast cancer patients could potentially be advanced by having a more complete understanding of breast cancer biology, including a catalog of recurrently altered genes. Over the last decade, thousands of human breast tumors have been profiled for gene expression and DNA copy number alterations, and ongoing efforts in DNA sequencing are establishing the set of somatically mutated genes. Much of the molecular data being generated resides in the public domain, available as a resource for further research. The challenge then becomes how to best utilize all these data, to mine them for candidate biomarkers and targets, which is the subject of this review. Some examples of integrative analysis and combining results from diverse data sets are also presented.