A strategy to classify human tumors into informative subtypes based on their mutations is presented in a paper published this week in Nature Methods.
Cancer is not just one disease, but a manifestation of many. Recent efforts to sequence cancer genomes are producing copious information about somatic mutations-those that are present in the tumor but not in healthy tissue of the same individual-in human tumors. Methods are needed that allow this information to be utilized to classify tumors into subtypes that give insight into the underlying biology or help in making useful predictions about the likely disease trajectory.
But tumor mutation profiles are often very heterogeneous. Individuals with a very similar disease may share one or few somatic mutations, and therefore using mutation data to classify tumors in an informative way is very challenging.
Trey Ideker and colleagues now show that knowledge of the network architecture of human cells can be used for this purpose. They reason that even if tumors do not share specific mutations, the parts of the network that are affected may be shared. By integrating somatic mutation profiles of human ovarian, uterine and lung tumors with network information from public sources, they show that network-based stratification identifies tumor subtypes that are predictive of patient survival and response to therapy.
COVID-19: Assessing instances of long COVID in UK health dataNature Communications
Health technology: New cost-effective smartphone test for middle ear functionCommunications Medicine