Integrative Bioinformatics Approaches to Discover Pathological Genetic Events Underlying the More Aggressive Luminal Breast Cancers
The Cancer Genome Atlas project has generated a daunting amount of genomic and deep sequencing data for tens of thousands of human tumors. This provided unprecedented opportunities to systematically analyze the cancer genomes to discover driving genetic alternations and develop novel therapeutics. In the past a few years, we have developed the computational approaches that interrogates multiple levels of genomic data to reveal cancer-causal genes and therapeutic targets.
Estrogen-receptor positive breast cancers can be classified into the luminal A and B subtypes. While the luminal A tumors can be effectively treated with endocrine therapy, the B subtype tumors are more aggressive and prone to develop early endocrine resistance. The genetic alterations underlying the B subtype tumors are poorly understood. In our research, we have focused on discovering the pathological genomic rearrangements generating recurrent gene fusions or gene amplifications that specifically promote the luminal B tumors.
In this colloquium, I will discuss our Fusion-Zoom pipeline that interrogates RNA sequencing, copy number, and concept signature datasets to reveal pathological gene fusions, and our identification of a recurrent ESR1-CCDC170 gene fusion that is preferentially present in the luminal B breast cancers. In the following, I will introduce our concept signature analysis and its application in revealing the oncogenes targeted by genomic amplifications, and our studies on a cell cycle kinase target frequently amplified in the luminal B breast cancer.