TUESDAY, Dec. 16 (HealthDay News) -- Examining subnetworks of genetic activity in a patient's tumor better predicts the spread of breast cancer than conventional techniques, researchers say.
University of California at San Diego scientists, working with Korean researchers, used bioinformatic algorithms to identify these subnetworks. They then mapped the gene activity to the many networks of signaling pathways and protein complexes that prior research had found in human cells.
The scientists explained that conventional "rapid microarray" technology allows cancers to be classified by gene expression, or activity patterns, but this system is imprecise because cells from a single tumor sample can have genes switched on in some cells (from one part of the tumor) that are inactive elsewhere.
The researchers' new work enabled them to identify subnetworks in which aggregate gene expression patterns can distinguish between patient groups.
They also uncovered many genes associated with breast cancer that had not been identified by previous gene microarray profiles.
The researcher team is now applying its analysis system to other cancers. These insights could give doctors a new tool for diagnosis and prognosis, they say.
The findings are scheduled to be presented Monday at the American Society for Cell Biology's annual meeting, in San Francisco.
More information
The American Cancer Society has more about breast cancer.
