A team of investigators identified networks of genes from breast cancer patients that they say are more accurate and reproducible predictors of whether a tumor is likely to spread.
The research used the same gene expression data in two prior studies from van de Vijver et al. reported in Nature and Wang et al. published in The Lancet. Each study yielded a set of about 70 individual gene markers that are now used in hospitals to help predict the likelihood of breast cancer metastasis, according to the scientists.
Using a mathematical approach for the prediction of metastasis, the researchers calculated the average behavior of subnetworks of proteins. “We saw about a nine percent increase in metastasis prediction accuracy over the two main sets of individual gene markers,” says Trey Ideker University of California, San Diego bioengineering professor.
He says that his team raised metastatic prediction accuracy for breast cancer to roughly 72%. The subnetwork markers also are significantly more reproducible between data sets than individual marker genes selected without network information, according to the scientitsts.
The investigators uncovered 149 discriminative subnetworks consisting of 618 genes from the patients from the van de Vijver et al. data set and 243 discriminative subnetworks with 906 genes from the Wang et al. data set.
Each subnetwork is suggestive of a distinct functional pathway or complex, the research team points out. For example, they showed that P53 plays a central role in several protein subnetworks—it interconnects many expression-responsive genes. P53 itself, however, does not show up as significant in conventional expression clustering or classification methods, the scientists report.
A compendium including all of these subnetworks is available online via the CellCircuits database, which provides each subnetwork in both graphical and machine-readable formats.
The study was conducted at the University of California, San Diego and the Korea Advanced Institute of Science and Technology. It is published online in Molecular Systems Biology.