Hundreds of oncogenes have been identified in the human genome through genetic mutations. Most oncogenes, however, are found in people with a particular type of cancer. Jess Mar, an assistant professor at the Australian institute of bioengineering and nanotechnology, says these genes are not the whole story. He believes that gene activity is different in patients of the same or subspecies, and that it is difficult to compare the level of gene activity in a group of patients by looking at data on the molecular makeup of cancer.
A new statistical method, called Oncomix, identifies oncogene candidates (OCs) in RNA-seq data. The principle of this method is to identify OCs by detecting low expression in normal tissues and overexpression in patient tumor subsets. Patients were grouped according to the expression of OC. To demonstrate the usefulness of Oncomix, Dr Mar and his colleagues applied the Oncomix approach to RNA-seq data from the cancer genome atlas (TCGA) breast cancer cohort and identified a set of five OCs (CBX2, NELL2, EPYC, SLC24A2 and LAG3). Chromobox 2 (CBX2) was associated with poor clinical outcomes, according to calculations and experimental evidence, and Mar suggested that this may be a driving factor for breast cancer and should be further explored as a potential drug target for invasive breast cancer. This finding underscores the value of the Oncomix approach. Finding hidden, specific oncogenes could lead to new treatments for cancer.